Playing with the internet of things – my new Twine

By March 11, 2013 5 Comments

My Twine arrived today and I’ve spent the last couple of hours trying to set it up to monitor how often our cat Polly goes in and out of the house.

For those that haven’t come across it Twine is a small box that  you can attach sensors to which then monitor things that go on in the real world. It can be temperature, dampness (to detect flooding), vibration, doors opening or closing, any manner of things in fact. The box is programmable via a web interface which lets you set up rules to email, text, or send an HTTP request when the sensor triggers.

Polly seems to be a very lazy cat (sweet, but lazy) and we often wonder how often she goes out, particularly at night. My thought when ordering the Twine was that it would be fun for the kids to plot her movements on a bar chart. The Twine sells itself on being dead easy to set up and do things with, so I thought this might be a reasonable ambition. I still think that might be doable by triggering an HTTP request every time the cat flap opens and then using an analytics service to plot the traffic, but simply getting the Twine working turned out to be quite involved so I decided to downgrade my ambition get started with an IFTTT based rule that creates a tweet on my brisbourne1234 account every time the cat flap opens and I get an email from the Twine.

I almost got it working. I had all the steps in the chain worked individually – the sensor trips when the cat flap moves, Twine sends email when the sensor trips, IFTTT rule sends a Tweet when my gmail gets the email from Twine. But it all fell apart when I was trying to string it together. I was lying on my back by the cat flap trying to Selotape the sensor to the side of the cat flap in a fixed position that would trigger every time, when the Twine lost connection with the dashboard making it impossible for me to see if the sensor was in the right place.

I searched all the forums and tried everything I could think of to get the connection back, changing the batteries, trying AC power, trying to set it up again, and I did come back eventually, but it’s still not updating frequently enough to properly test whether it’s working. I’ve just moved the whole arrangement to the fridge door where it is easier to fix the sensor in a position that looks like it will work and I’m going to cross my fingers call it a day now.

I think that maybe a slower moving target like the fridge door would have been better to have started with and that the cat flap might have been a bit ambitious anyway.

What do I think about all this?

I think I’ve had a glimpse of the future and it was quite fun whilst I was making progress trying to get the Twine to work. But it ended in frustration and I think we are going to have to step up a couple of levels in usability before a gadget like this goes mainstream. Additionally, the fact that monitoring the cat flap is the best thing I can think of to do with it suggests some work on use cases is in order….