Trying to catch up with what's been happening on the web whilst I've been too busy to blog...
The first thing I want to point you at is part of Matt Webb's recent presentation to Web Directions North 2008. The main thrust is about how we could extend RSS readers to let you process to-do list items more quickly, and is something I've thought about off and on since he and I exchanged a couple of emails when he first mooted the concept he's now named SNAP. As he's found, it's not quite ready for primetime, so it'll be a while before we introduce it to tedium. However, some of the other concepts (like the state machine for GTD) fit quite well with things we are working on, and it's nice to have a way of explaining it.
But the main reason that this isn't just a del.icio.us link is because of the future challenges for the web that he lays out at the end of the talk. In particular, the slides about groups.
I think designing software that's used by groups rather than individuals is going to become an increasing problem as computing escapes from computers and into other devices. I don't have any decent answers, but it's something I have thought about a bit. It's obvious that you don't want people to have to log in before they can use their radio, so I think the only other option is to have the radio notice who's around.
The easiest way to achieve that today would be for it to track Bluetooth devices - if my phone is in the room, there's a fair chance that I'll be in the room too. The only problem with that is that by encouraging us all to wander around with Bluetooth turned on we're also encouraging everyone to open themselves up to being tracked everywhere. Maybe our phones need to become cleverer about who they reveal their presence to, but then we've just shifted the difficult interaction from the radio to the phone...
I've been thinking about getting paid. I was going to say I'd been thinking about revenue models, but that's just a fancy way of saying getting paid, which is maybe part of the problem.
Basically, the web has promises to be the best way for people to connect to each other, and so should be the perfect marketplace. The long tail promises that there are an almost unlimited set of niches waiting to be filled.
In parts, this is true. Take the process that sparked this blog post for example. This morning, Russell Davies wrote on his blog that he'd like a twitter feed of changes in the Amazon sales rank of his book. The idea piqued my curiosity, and an hour or so of poking round the web later, I've found out how to query the Amazon API to get the sales rank for his book and found out how I would hook into twitter to submit the updates.
Now, if I was just knocking something up for myself, I'd be about half-an-hour from having it all finished. But as Russell says, it's the sort of thing that other authors would find useful, so I've been pondering making it into a full-blown service.
"Making it into a full-blown service" is a bit over-the-top, but as Eric Sink said, it's a non-trivial step to go from something hacked together for me to something that I'd be happy letting other people use. There'd need to be a way to sign up, a way to stop the updates, and then the ongoing maintenance if either the twitter or Amazon APIs change.
It still wouldn't be a grand undertaking, but it becomes more like a day's-worth of work now, and then an undefined amount more in the future (but again, probably not too much). It's at this point that it stops looking like a fun problem to spend a while solving, and more like work. I don't want to launch services that I can't maintain, and obviously there's a limit to the number of services I can maintain - particularly if they're being maintained in my spare time.
What the web is missing is an easy way to charge for such small, niche services. Surely something like this is worth the price of a cup of coffee to authors? The problem is that, at present, the assumption on the web is that it should somehow be paid for by advertising, which means that the only things which get built are either a by-product of delivering audiences to advertisers, or things that geeks build for themselves.
If there was a way for people like me to cover their costs (plus a little extra) then we could solve all sorts of niche computer problems for people who can't code, without having to spend all our time working out how to force them to click on adverts.