I was inspired to write this article on a train ride back in February, and that's where I wrote most of it, too, without any internet to do any research. So of course, as soon as I plugged in I found articles like this one from more than two years ago, and I'm once again reminded that there's not an original thought in my head. But, I already wrote these 1,000 words, so here they are:
In my day, I've owned PDAs, mp3 players, a few Game Boys, a digital camera, and a cell phone. These days, I only carry my cell phone, my laptop, and my iPod with me regularly. The bulk of these devices is a reason that I stopped carrying some of them regularly, but it's not the primary reason.
At one point, I had wireless internet on my Palm V thanks to OmniSky (are they dead yet?). It was nice for what it was, but even then it seemed obvious to me that that was merely a temporary solution. Today, convergence is the name of the game, with more and more cell phones providing PDA functionality, and some are even providing gaming functionality.
It seems fairly obvious to me that the mobile devices (including cell phones, PDAs, game players, and to a lesser extent mp3 players and cheap digital cameras) are rapidly approaching a convergence. I won't be at all surprised if the more popular devices are going to be the ones that provide more features.
For that matter, the home entertainment market is hurtling towards convergence, too, with my PS2 able to play DVDs and connect to the internet.
Nintendo currently dominates the portable gaming market, with its Game Boy line being the only real contender. However, Nintendo has said on multiple occasions that they are a game company, and were only interested in supporting gaming with their devices. Witness the lack of DVD functionality in the Gamecube as a prime example of this. Generalizing this behavior and putting it in terms of what I've been talking about, it's unlikely that Nintendo is going to be very receptive to the convergence trend. In fact, I would be honestly surprised if Nintendo hopped on that bandwagon, and bundled phone or other features into the base Game Boy configuration. Sure, Nintendo may reintroduce peripherals like the Game Boy Camera or the Game Boy cell phone connector, but accessorizing is more Nintendo's habit than converging. (Because there's more money in accessories!)
Meanwhile, I wouldn't be surprised if Sony and Microsoft (and others) attempt to compete in the portable gaming space not with a dedicated game system, but rather, with a gaming platform piggybacked onto other devices, with the most likely target being cell phones. Heck, Nokia's upcoming N-Gage is just the first in what I expect to be a string of gaming platforms piggybacked on top of other devices. And I think, with that strategy, it is likely that someone will overtake Nintendo's dominance in the portable arena, primarily because I don't expect Nintendo to embrace the convergence that their future competitors will be offering. The more popular devices will be those that offer more features, so people will buy phones that also happen to be able to play games, and since their phones support the games, they'll buy the games for that platform.
So convergence of portable devices is coming, and I expect piggybacked platforms to have a pretty significant impact, but I'm not actually happy about that. And it's not just because I'm a Nintendo fanboy. No, I'm not happy about this because I subscribe to the philosophy of "do one thing and do it well." Witness my love of UNIX as proof of this. So I'm not happy about this convergence that I'm expecting because I'm afraid that we're going to end up with a sea of devices that do many things half assed. My iPod plays music incredibly well. My Game Boy plays games well. My phone answers calls well. My CellularGamePod, on the other hand...
I don't mind convergence. It's not a bad thing, and for people who can't afford multiple devices, it's probably a good thing. But I think a better solution that could be more significant than convergence in the long run is the idea of a "Personal Area Network."
- Components of a "Personal Area Network"
- A cellular link node for network/internet access. This could be a cell phone, a computer, or even a PDA with a wi-fi card installed.
- Bluetooth (or another, ideally less "nasty" protocol) to allow devices to communicate with each other. This would also enable network access for non-networked nodes in the network.
- An auto-discovery protocol. Preferably Rendevous, but (if I recall correctly) JINI might also do the trick.
- Some standard protocols for sharing standard data between devices.
- Then, add devices to yourself as you see fit. Some of these devices could be devices which offer more than one service.
- And most importantly, all of your data is stored on your local devices -- not accessed through the network.
The advantages and uses of such a PAN innumerable. You could access the contact list on your iPod with a borrowed cell phone to make a quick phone call. You could connect your laptop to the net using your cell phone or a friend's. Your Gamecube and Game Boy could talk to each other for connectivity. Two Game Boy owners could play each other without need for a cable. Take a digital picture and email it to a friend through your cell phone, or store the picture on your camera as identification, without using a shitty camera-phone. And on.
Basically, some devices would "offer" services, and other devices would "request" services. Phones both offer and request contact data. My Game Boy requests net. My laptop would offer and request a lot, depending on its current configuration (I'd certainly want my laptop to offer net if it were plugged into a cable modem, instead of messing around with a cellular link). And my iPod could offer music and conacts (though bluetooth doesn't have the bandwidth to stream music -- just one of many reasons this would ideally use a better protocol than bluetooth).
When my phone connects to a device offering contact data, it would browse that data remotely and not save a local copy, unless the owner of the data allowed it (though if the owner doesn't want you to download the data, he probably wouldn't let you browse it in the first place) and the browser of the data requested a copy of the data. This protects the owner of the data, and keeps each device from getting cluttered with data unnecessarilly.
I probably shouldn't be emphasizing contacts so much, because that's an example of pure data that should arguably be availble just from "the network", but that's a topic and an argument for another article. Better applications for this PAN technology are probably things like the digital camera services and the net services I described above.
The biggest unsolved problem in the sketch of a PAN I described above is security, but I can imagine numerous possible solutions to that problem. The first solution that popped into my mind was giving each piece of hardware a hardware based keypair. Don't make any way to get the "private" key from the hardware short of tearing the hardware apart and working some EE magic. Then make a "known_hosts" for your PAN, and your PAN can use something akin to ssh host key verification to allow access to your network. Of course, firmware upgrades would always have to be supported in case of flaws in the protocol... Something a little better than having to manage a "known_hosts" would be cool, but like I said, this is just one solution of many. And heck, for all I know, bluetooth might already have some way to do secure ad hoc networks, but I'm on a train right now without a PAN, so I can't connect my laptop to my phone and check. =)
I'm sure most of these ideas aren't very novel, but the entire vision is still my ideal future of digital devices. I wish I could do more than just hope things turn out this way.