Miker

17th level Hacker

Mobile Data Services and Innovation

Lots of times I groan at the statements that come out of the MIT Media Lab, but TheFeature has an interview with William Mitchell, director of the MIT Media Lab and I really liked what he had to say at the end of the interview.

We must understand the relationship between these rather abstract technologies and our everyday lives. We need to ask ourselves how we want to live our lives and how we can organize our technological capabilities to fulfill those ideals. Public debate is critical. And you can only have that debate if you engage the public’s imagination.

Rock on, I think that’s a great attitude for potential innovators to have. Lots of times us technologists run off and start working with new technologies and forget that they eventually have to be used by people who might have no idea what we’re talking about. Especially in places where many of the people you interact with every day are at least somewhat technically adept (such as here in Silicon Valley or in the labs at MIT) it can be easy to lose perspective.

Realizing that the pitfall exists is only part of the problem however. As William says, you need to engage the public in a dialog about what needs they would like served, and in order to do that you need their attention and you have to capture their imagination. Well, for most people they aren’t very interested in a new technology unless it’s something they can use already. They don’t really want to play with demos or prototypes, they’re concerned with what they can do to solve their problems. So you have a kind of chicken and the egg problem of users needing to drive the development process but not being really interested until a product is ready. One of the traditional solutions to this issue is focusing on a group called the “early adopters”. These are the people who are willing to put some time into finding the correct solution for their problems. They’re a little more experimental and are normally highly motivated to establish a new solution to their pain points. What a tech company will do is try to service that early group as best as they can, and try to use those people as a proving ground and to collect input. The hope is that the early adopters can provide some insight into what needs to be done to service the main body of users who will hopefully be following some time later. “Some time” equals however long it takes for the product to reach a stage where the user no longer has to put up with quirks and shortcomings in the system, they just get their packaged solution at a price which they think is justified by the value they’ve been provided.

The issue that I have pretty much constantly with “mobile services” is that they make it difficult to run these early stage experiments. Many of the wireless network specific features (SMS, EMS, MMS, WAP push, geolocation information, etc) aren’t presented in a way that makes it easy for a third party application provider to use. Say for instance I’m working on something in the style of the Home of the Future referenced in that article. Say I’m making an Internet appliance that I want to sell to people that they can hook up to their broadband connection in their house. This isn’t a big deal, broadband is apparently really gaining in popularity both here in the US and abroad. It used to be that providing a service which required your standard home user to get a DSL or cable modem could be a pain. But this is no longer the barrier that it used to be. I want the device to provide a web interface, and I want it to be hooked up to cameras, control equipment, and sensors. Also, not a big deal here. There are lots of people with IP cameras or simple webcams. There’s a hell of a lot of control and security hardware meant to be hooked together and fed into computer control systems. This also isn’t hard to get working.

But now say I want that system to be able to interact with the owners wireless device. And I don’t just mean providing a WAP page they can use to check on their system, that’s really just the most basic level. What if I want to send those images from the camera as MMS if the user has gone away on vacation and someone opens the front door? Well, now we have some issues. All of a sudden our well connected world and it’s seamless information architecture suddenly seems very discontinuous. How do we get a message from the Internet to a mobile device? Well, the answer is normally that it’s just not that simple. Unless you want to involve the cell phone service providers whom your customers for this Internet appliance might be using, you’re pretty much stuck. So that means going to everyone (ATT, Cingular, TMobile, Sprint, Verizon, Boost, Orange, TIM, Vodaphone, Wind, and on and on) and making some deal with them so that you can provide your Internet appliance to talk to a mobile appliance. Or you could restrict yourself to just a few of those providers, and selling your appliance through channels that select only users of that service. And of course, in order to talk to people like Cingular and Sprint without getting laughed out of their offices you need to have a very solid revenue stream in place you hope to share with them. They normally aren’t interested in your great idea unless it’ll make them money right away. Not a great environment for innovation.

So I agree that we need to engage potential users in a discussion about how to best service their needs. But I’m very sketchy on how we provide the prototype systems and trial deployments necessary to foster that discussion without bypassing the existing wide area wireless infrastructure. One option is to just ignore the enhanced services of the wireless networks, and just use the data capabilities to build what amount to standard Internet applications. But there are lots of issues with a model like this, many of them too technical to get into here I think. What we need is the ability to experiment with new services before they’re finalized, otherwise the applications that get forced into the wireless pipe are going to be immature and probably poorly positioned in terms of value. I think we’re already seeing this happen to a degree. MMS hasn’t seen the uptake that most were expecting it to. I place the blame squarely on the MMS infrastructure being more restrictive than that of SMS. It’s hard to know who you will and will not be able to send MMS to, and exchange agreements between service providers are ill defined if they exist at all. Normal users aren’t going to bother taking the time to learn a new service if they aren’t going to be able to use it in all the time. Technical restrictions like service provider peering agreements are indistinguishable plain old failure if the end result for the user is inability to complete their task. Even worse because the failures tend to be silent.

How do we come up with a system that allows this experimentation? I have no idea. But I’m fairly certain that someone needs to work out how to apply better end-to-end design with relation to hooking Internet services up to mobile services. There are issues.. I know there are issues with trying to do this. But that doesn’t mean it’s not the right thing to do. With all the talk about the dissolving barriers of communication, and electronic information flows, I’m surprised to see what seems to be an increasing gulf between Internet and mobile services. The flow of bits from my home network to my cell phone and vice versa appears to be anything but seamless.