I've been having a lot of fun with the ESP8266 lately and exploring the various consequences of being able to wire up arbitrary things with Wifi for very little money. One wild idea that popped into my head is somewhere between Geocaching and a public art project, which I call a Sound Capsule.
A Sound Capsule is a public Wifi network, but when you connect to it there's only one webpage you can reach (like those awful airport wifi networks). The page has a little audio player on it that plays some sound, and a button that lets you record your own audio for the next person to listen to. However, you can only record one thing at a time; the previous one gets deleted. The device itself would be a self-contained solar/battery type rig so you could hide the physical components away somewhere safe; you only want people to interact with it as a mysterious Wifi network.
I think it could be pretty interesting seeing what people do with it. Would it just be all random trolling or would people actually leave interesting messages? Maybe you could have a whole conversation with a stranger. Or maybe people would just make silly noises. It really feels like it could go either way.
Things got a little tricky since Monday. I've been particularly busy lately, and I think adjusting back to longer posts is something I could have prepared for a bit better. My writing mainly runs on habit, and shaking up that habit was perhaps dangerous. To top it all off, normally I'd be writing in advance which would absorb a bit of variation, but I've fallen out of doing that because of technical issues that meant the posts weren't appearing when I posted them early.
I think in this case the main kind of preventable failure is not heading off the near misses and various warning signs early enough. Actually, I wrote about grading things on a "fail scale" some time ago, which I neglected to do for this. If I want my writing to be consistent it's not enough that it merely happens, it needs to happen comfortably. That way when things go wrong I'll still have some margin to work with.
With this I conclude my week-long experiment in brevity. I can't say I particularly enjoyed it, but it was definitely worth trying. One observation in particular: it's easy to assume, especially for computer folks, that you can just change the underlying implementation and keep the high-level behaviour the same. But it's just not true. Expressing certain things is easier in certain formats, and changing the format changes the kinds of things you express. For what I want, I think longer (but not too long!) is often (but not always!) better.
A great idea came up the other day: a Make it Rain app. Instead of giving someone physical money, or making a boring online transfer, you could point Make it Rain at them and flick virtual bills at their face up to the value you want to send. The transfer itself could be backed by PayPal, Bitcoin or similar, so you can use it without the other party having to install anything. However, if you do receive money with the app open there would obviously be a flurry of bills tumbling down the screen. What a time to be alive.
A funny thing I've noticed is that, although I really dislike stress, I work very well under pressure. When the hammer drops and there's a looming deadline, it's easy to keep moving, keep making decisions and do the best I can under the circumstances. This seemed like a paradox, until I recently realised that the difference has a lot to do with practice vs performance. In a high pressure situation it's difficult to think, but easy to act. So maybe it isn't so much a preference as a balance; low pressure for creativity, high pressure for output.