I’ve got a list of stuff I would like to read up on and play with, and NFC has been on that list for a while. I’ve been really excited about the secondary stuff that should go along with NFC. For a long time folks inside the industry have been vocal about NFC not being “just about replacing your credit cards with your phone”. Theoretically NFC should also open up all kinds of interaction with the real world. We had a Mobile Monday Silicon Valley panel on NFC last night to talk about some of the overall issues. I have to admit however that I feel like the whole thing is headed straight for a brick wall at high speed. Even though I would love for it to work, it’s not going to work.
NFC as it stands is setup to suffer from exactly the same set of issues that has effectively killed Bluetooth as everything except a wire replacement protocol. The problem is that when NFC pushers say that NFC is “about more than just replacing your credit card with your phone” what they really mean is that NFC could be about more than just replacing a credit card. The decision about how capable or restricted an NFC stack is lies with the equipment manufacturer. And generally speaking folks don’t just toss random features into phones because they’re available. They do so when it’s really necessary, normally because of consumer demand.
Right now the NFC field is going through something very similar to what the GPS world went through. I’m sure some of you will remember that back in the deep dark recesses of pre-history not every smartphone had a GPS chipset in it. Those of us geeky enough to care went out of our way to find the models that did have GPS support so that we could play around with our own little hacks. But back in those days the consumer public didn’t really care about GPS support in phones. Back in those days the only folks who were clamoring for GPS support in every handset were the folks who were selling navigation or mapping apps for phones, and they were charging way too much, and not sharing revenues back with the manufacturers. So the environment was pretty well jammed up.
Eventually Google Maps came around though. And a meaningful free mapping application with fantastic search changed the potential value of having a GPS chipset in a handset, made the handset more attractive to customers across the board, resulted in more sales for handsets with GPS even if the manufacturer couldn’t sell additional GPS related apps, and we ended up where we are today. Generally that pattern is true, to get a new technology out into peoples hands there has to be a killer application to drive it. And the version of the technology that gets deployed will be the minimum version required in order to make that kill application work, and that ends up being the new standard.
Contrast the GPS uptake with the way Bluetooth has worked out. The Bluetooth technology folks early on used examples ranging from the headset connection version everyone knows through syncing all your data between devices automatically. There’s a whole ton of potential functionality in the Bluetooth specs, and at this point it probably is possible to sync data wirelessly and automatically in some subset of devices. However, the killer app there was connecting a wireless headset, so that’s all that anyone really uses Bluetooth for these days. And on lots of devices that’s all you can do. I’m sure that statement is going to piss of a whole industry. But tough, it’s true, deal with it.
With NFC the killer application appears to be payments. So my pressing fear is that the version of NFC that gets deployed to the devices out there is going to support the subset of features required to be able to participate in payments, and that’s it. Now, I know, the subset of stuff required to support payments potentially covers a few other areas like being able to scan other styles of tags. While I understand some of the convenience points there from a consumer perspective, there are some massive downsides from the point of view of an independent app developer. MASSIVE! Having a fully functional system requires a new bit of hardware even if I have an NFC enabled phone. That’s a huge barrier to experimentation and uptake for anyone except the folks with deep pockets and long timelines.
I was hoping that after the discussion last night I would find out a few things I was unaware of. I would be able to grab my Nexus S and with some quick bits of hackery potentially do something interesting. While I admit that NFC could potentially deliver some wins for a subset of cases out there, it’s not something that an independent app developer needs to take into consideration when building applications. I’m hoping that things shift somewhat as this evolves, but as it stands I’m taking NFC off my list of stuff to find some time to play around with. I just don’t see any areas in which it’s close enough to being a practical component of mobile development for someone trying to build a new business in mobile. Those of us looking for additional ways to make applications more context aware are going to have to stick with other techniques for now.