There’s an article at textually.org, SMS ‘Smart Bracelet’ for Medical Emergencies, which made me think of the Wherify GPS watch and the planned project called Wheels of Zeus launched by Steve Wozniak. The world certainly seems to be converging on a “sensor network” view of computing, in terms of pervasive devices using wireless networking, sensors, and geolocation information in order to carry out tasks. These ideas have been around for quite a long time. I remember hearing the starry eyed pundits discussing Internet enabling just about everything back when I was in college. Most people laughed at the time, other people scrambled to figure out how to get wireless interfaces and sensors into just about everything they could get their hands on.

I’m a big fan of technology like this, and I’ve been involved in a few ventures aimed at internet enabling various technologies. So I would like to take a few moments to bring up the counterpoints I’ve encountered along the way.

The first issue is mostly an issue here in the US - cellular wireless is expensive and highly incompatible. Cell phone service providers might be shaping up now that users are allowed to move from one provider to another without giving up their phone number, but I wouldn’t count on too much. And the common advice handed to the cellular service companies by advisors is that they should be reaching up higher in the value chain in order to recoup their investments on infrastructure. That means they won’t be making it much easier to deploy applications without involving the service providers. Anyone who has tried to deploy applications that use SMS and MMS services should know what I mean. You have to talk to all the service provider companies that any of your users might be attached to in order to find out how to deliver messages into their networks. There is no basic equivalence between the Internet and cellular networks, and trying to build applications that span both can be difficult. Keep this in mind when looking to determine how to get geolocation information. Although the cell companies might have it in order to comply with enhanced 911 requirements, that doesn’t mean that your application will be able to use that data.

The second issue is privacy. This one has gotten a decent amount of attention, so I won’t harp on it too hard. But having your location information available means that others can find you. Pretty obvious. But it can be a very slippery slope. It’s very easy for today’s helpful child tracking application to turn into tomorrow’s Big Brother scare. Just look at the whole overblown scare about the use of RFID. If RFID debacle has taught us anything, it’s that there doesn’t even have to be much merit to claims of privacy concerns once the ball gets rolling.

The final point I would like to raise is access to information from mobile devices. For the full vision of these pervasive mobile services to be realized, users have to be able to access this information all the time. Yet deploying mobile applications is still painful. There are lots of mobile platforms, lots of competing standards to utilize and to try to comply with, and in general the devices are still constrained. What has given mobile computing a shot in the arm as of late has been 802.11 and wireless laptops. I personally think this accounts of much of the new resurgence of interest in mobile computing among the general populace. But those looking to deploy applications which will be accessed from smartphones or PDAs will find much of the same muck that existed 5 years ago, before everyone started laughing at the meer mention of using WAP. Cross platform applications targeted at mobile devices are just as difficult to build now as they were then. It can be done, but it will take a lot of effort. Don’t buy into the hype that the mobile computing world has changed, it really hasn’t. Don’t underestimate the effort required if you’re looking to mobilize an application.

So those are my words of caution on that note. There is some cool new technology in the pipe for 2004. But for once, I would like to see applications that actually work deployed to end users, and products that are still platforms for rapid innovation not oversold too early in their lifecycle. The Big Bad Business Pundits seem to think that innovation and business can live side by side, and us technology wonks can certainly do more than we have been in order to help realize that goal. Maybe I’ll make that my New Years resolution.