@pablo :a lot of this hype, true. But today it's much easier to develop and prototype ideas about IOT and we do have new technologies to add to the mix ,so many more ideas get tried by people of varied backgrounds and some will have good value.
I am very skeptical of the IoT hype. Most of the applications people mention for IoT have been possible and I would argue practical for decades. I have been in the home automation industry for a long time and I have seen many technologies arise which made similar promises only to settle into niche markets or fade away: X10, CEBus, Zigbee, Z-Wave, UPnP, and so on.
I can only see two differences with IoT:
1) the current ubiquity of smart phones
2) more than the usual amount of hype (engendered mainly by #1).
So if the smart phone isn't the driver for IoT, why would people suddenly start wanting intercontinental lighting control and big data infographics of their refrigerator contents?
Weary or not, I don't think you actually read my comment. The interesting question is how IoT is going to develop the interoperable protocols necessary to make it work, work *together*, in useful ways. The current path seems to be "throw all data to vendor-specific cloud sites and hope that someone like Google saves the day by making interop".
It's strange that the industry has gotten away from classic IETF-like approaches - I speculate this is because lawyer/MBA types are constantly braying about IP.
I always get weary when reading about exactly what the Internet of Things will be, or will not be, or has to be. As if anyone could really know at this time.
As others have pointed out: equating IoT with wearables appears as a rather narrow view. @junko.yoshida makes a good point of that (although meaning the opposite): all the examples appear to be non-wearable, not even mobile ("on the wall" etc.).
One mistake is that the author seems to equate wearables with IoT - not so! There's nothing wrong with activity trackers, but there's so much more to IoT. Garage-door sensors, spoiled-milk detectors, smart toilets, fire alarms, smart water faucets, pet foot sensors, whatever.
The other big problem (not limited to this article) that IoT works by connecting everything to the cloud. That's just crazy - or rather, it's just sloppy thinking. Yes, the cloud is handy for a certain level of connectivity, but we don't want everything going there. For privacy reasons, as well as energy ones - what we need is *protocols* to permit local handling of IoT data.
Protocols and standards (ie, public, vendor-independent ones) are what needs the most work.
When the author said, "For starters, if the IoT market is to reach its full potential, the tether to the smartphone has to be cut," he clarified what he envisions as the quintessential IoT device.
I tend to agree with him. The IoT device, in my mind, is an unintrusive device that people don't even notice that it's there, but in fact, it's everywhere. It's on the wall, it's on the floor, it's on the door, it's on streets. But probably not the smartphone, with which people have just too many interactions for other purposes.
Rick, the IoT devices can definitely not afford to build their own LTE connection. However, they can have Low Power Bluetooth, as an example. Right now, the components and software are not quite there. Although they are on the horizon. Not just on a wearables side of the equation. We've seen more and more BTLE hubs as stand alone devices and being added to existing wireless routers.
If the application, for whatever reason, necessitates an interaction with a cellular network, then allowing the wearable to capture data unteathered for a time and then uploading the data when reconnected to a phone or a BTLE hub is certainly one alternative. Some applications, like in the jogging hypothetical that AZskibum referenced, may need real time communications. In that case, relying on a gps chip on a phone isn't ideal. The wearable itself is going to need low power sensors.
The radio is just one variable in the untethered equation. The low power sensors and flexible batteries or alternative power sources also need to be developed. Granted, the technology to do these things doesn't exist now. Don't blink! It is coming.
"People simply don't want to take their phones, especially large form-factor smartphones, with them everywhere they go."
Or perhaps just the opposite. We all know people who can't imagine going anywhere without their smartphone.
For cost & power reasons, today's wearables primarily use BTLE to upload their stored data, usually to a smartphone. Why not leverage that LTE link & apps processor that is usually "worn" at the same time as the wearable device. For those cases where the user prefers to not carry the phone (while jogging, for example), is it such a burden for the wearables to retain it's data until it is once again within range of an internet-connected device? How many consumers would pay extra for a wearable that has it's own independent internet connection?
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...