Tom Morris

I popped out for about four hours today to do about an hour's worth of things. That's because the only practical bus is one which goes once a day in each direction. I curse the beast that is not having passed my driving test yet (and not having enough cash for the extravagant insurance rates nor the equally extravagant cost for driving lessons). Public transport can work. But it doesn't work here!

That said, this morning I've saved about £150 on travel costs. I've found the most badass price for a flight.

Comments | TrackBack

I saw this story today about mobile phones for the elderly. Cool idea, especially the way it works with hearing aids. Here are some other thoughts about phones for the elderly, and especially about "alarm buttons".

Make the ringer loud. And make it so that if it's a fixed phone, the noise is directional. My gran used to not hear the phone ring, despite the fact that she had an extra bell attached. Direct that sound. Make it so that if you've got a fixed phone, you can have a radio ringer that plugs in to an ordinary plug socket. That way, you can have the phone in one room, and have a ringer right next to the person's chair in another room.

The "red emergency button" is also a great idea. Here's what you need - a preset logic to it. One of the main frustrations was that we wanted to get one of these for gran, but they cost a lot to run. The phone company charged a monthly fee for the service, and you had to have one person who was there in case of emergency.

Here's how you could route around the problem. Have it so that it uses three networks - the mobile network (rather, all mobile networks), the standard phone network (have a little adaptor that plugs in to the socket and have a spare) and have it work over VoIP - with wi-fi. And prioritise that crap!

It's all well and good using the mobile network, but it needs to be failsafe. You only need to slip and fall somewhere without mobile service and you're screwed. Ideally, mobile and VoIP networks would make these emergency calls totally free.

Build some logic in to the process so that it can try numerous people. Part of the problem with current "alert" buttons is that they dial only one number, putting the responsibility clearly on that person. Make it location aware, so that if there's a panic at home, it calls the neighbours. Also, make it so that if it can't get a connection, it lets off a really loud noise to attract the attention of others. Make sure that if the neighbour is on holiday, it calls a different neighbour. Build a complex system that's easy to set up.

I'm imagining how cool it will be when we can have this setup to auto-trigger. A sensor built in to our body that can transmit the data to the phone upon major health failures (eg. heart attack, stroke) and automatically call the relevant people - ambulance, doctor, next of kin, building manager and so on.

The population is living longer. We need this stuff far more than we need more Java games on our phones. There's little competition in this market but lots of potential. If you're thinking of building something, build this as it has the potential to save hundreds or thousands of lives.

Comments | TrackBack

I haven't yet had a chance to buy or read The Long Tail, but has either Anderson or someone else done any work looking at the Long Tail in comparison to temporal distribution? I've just looked at my del.icio.us account and the tags seem to be distributed in a long tail form (though I haven't done the graph yet).

It would be interesting to see how one could add a second Y-axis which would be some time variable (what the variable would be could vary - for weblogs, it could be "average amount of time between updates", "number of updates per time x"). Other possible metrics that one could use could be things like number of words, average sentence length, average word length, average paragraph length, amount of words to amount of pictures (in square pixels) ratio, number of links per post (or footnotes for offline), reading grade and readability metrics and - how about this, eh? - number of HTML/CSS (etc.) errors.

When I'm less busy, I'll put some of these charts together.

Comments | TrackBack