Early morning of Day Two of CES, and I see a problem emerging. I have been asked five times in the last two hours “What I saw at the show?”, and my answer so far has been “I have not actually been inside yet.” I had a brief foray into the Central Hall. I know the Qualcomm booth is next to the Intel booth. I know that the Starbucks at the Las Vegas Convention Center Starbucks has been remodeled and that it no longer  has enough seats.

I spent all of day one in meetings. One truism of CES is that there are never enough places to sit down and have a conversation. I stood for half of my meetings yesterday. And even though everyone says the show is less crowded this year (because of cancelled flights and bad weather) CES is still very crowded.

Every year, the press seems to pick a theme for the show. The CEA, the body that organizes CES, seems to have some input on this, as do the big exhibitors. This year, that theme appears to be “The Internet of Things” (IoT) or its more consumer-friendly child “Wearables”. Readers of this site should be familiar with IoT, this is the new name for Machine to Machine (or M2M) systems, a subject on which I have written in the past. In that post from May I noted that M2M has been ‘just over the horizon’ for ten years, but finally seems to have some real momentum.

For those new to the subject, M2M or IoT, is basically a concept in which every piece of electronics or machinery has some way to communicate with the Internet. “Wi-Fi in your toaster”. The new-found momentum is driven by several factors. Radios and Wi-Fi have gotten cheap enough to embed in just about any form of electronics. In addition, the industry now has the software tools to actually make use of all the data that an M2M system throws off (aka ‘Big Data’).

However, there are still a few key pieces missing from IoT. First, there is still the very complicated problem of middleware, or the network that has to sit in between all those cheap radios and all those big data systems. How do you connect dozens or hundreds or thousands of remote devices? There are some interesting projects out there looking to tackle that problem, some of which I have written about.

The other problem is that no one really knows what devices consumers want to get connected. This problem exists for enterprises too, but that is a post for a different day. Exhibitors here have booths filled with interesting ideas. We have moved beyond connected alarm systems, to connected appliances – toasters, refrigerators, baby monitors, etc.

The industry have been down this path before. During the 1990’s Internet Bubble there was a lot of talk about refrigerators ordering milk when you ran out. The Internet-connected devices that emerged from that were pricey flops. This time around, the M2M applications seem much more practical and thought out. Nonetheless, there is still the risk that we run up against a wall of consumer apathy.

An example can make this more concrete. I want to install video cameras in my home for security. My alarm company and my cable company both offer this service, but they are very expensive, closed systems. Alternatively, I can go to Costco and buy a reasonable camera system, but then have to install it myself and figure out how to access it over the Internet. In both cases, the bottleneck is that middle layer which connects the cameras to the web. My choices are a closed, proprietary system or an overly-open DIY solution.  Too many choices, none of which are perfect. And so, for the time being, I am not buying anything. Better to wait a year.