Category Archives: Blog

As we’re nearing that great time of year when local farm stands (and neighbors’ backyard gardens) have local tomatoes, I couldn’t resist what appears to be an advertorial by Analog Devices (a.k.a. ADI) that appeared in a recent Technology Review. I mean, how could I pass up a piece entitled IoT: The Internet of Tomatoes?

ADI is headquartered in Massachusetts, and the piece starts out with a critique of New England native tomatoes. I went to grad school in Massachusetts, and I don’t recall the native tomatoes being awful, but I’ll take ADI’s word for New England’s tomatoes being “comparatively tasteless” fare that generally get turned into ketchup rather than get served up in a nice Caprese sandwich. Anyway, ADI partnered with some local farmers on their Internet of Tomatoes project.

“This precision agriculture experiment uses technologies such as micro-electromechanical systems (MEMS) and sensors to figure out whether environmental monitoring could improve flavor.”

It looks at two aspects of tomato-cultivation: temperature and “growing-degree days.” Their launch was in January. Having been in Massachusetts in January (not to mention having spent my life in Upstate New York), I can promise you that there’s not a lot of farming going on. But January 2015 – that was the year New England really got clobbered with a ton of snow – would have made a good test case for technology survival. Working on a farm that lacked both usable power and Internet access, they decided on wireless (Bluetooth) sensors that communicate with handhelds (iOS or Droid). Data being communicated includes average daily max and min temperatures that are specific to each farm. The system is also incorporating info from a pest-management application, so farmers can keep tabs on beetles. Overall, the goal is to figure out best to cultivate tomatoes, and the best time to harvest them.

ADI is also taking a look at the chemistry of tomatoes. In doing so they’ve figured out why New England tomatoes aren’t as tasty as those found elsewhere:

“They don’t have high fructose, glucose, or salt content—which is why when most people grab a New England–grown tomato, they also grab a salt shaker…From an aesthetic perspective, the New England tomatoes look like any other. But on the inside, the chemistry doesn’t match up, according to ADI’s analysis of thousands of tomatoes.”

ADI is also experimenting with a non-invasive way to test tomato chemistry, not just for taste but for nutritional value. They’re working on an optical system that uses nearfield infrared technology. “Optical technology will [also] measure the soil, leaf, and stem lengths and their stress levels.”

Overall, it looks like an excellent use case for the IoT.

But about those salt shakers: I was of the belief that part of the pleasure of eating a native, Upstate New York tomato is having the tomato in one hand and the salt shaker in the other. Given this, I’m guessing that ADI’s Internet of Tomatoes will find a home in my neck of the woods, and not just in New England.

We’re all familiar with Underwriters Lab, and we’re used to seeing the UL symbol on the cord tag of our electrical appliances.  But things electrical – which were just being introduced when UL was founded in 1894 – have given way to things electronic.  And (depending on who you’re listening to) sooner or later, everything electronic (including all those UL-labeled appliances) will end up as part of the Internet of Things (IoT).

Earlier in the year, the UL published a cybersecurity standard “for the testing and certification of connection devices.” UL 2900 aims to address what has been one of the more significant barriers to wider spread IoT adoption: security.

While some have welcomed UL jumping in with their Cyber Assurance Program – they believe that it’s useful to have a third party certification available in this arena –  there’s concern that there are already so many security requirements established by vertical industry segment, that there’s really no room to adopt a truly common set of standards.

EE Times last week published in interesting article on this, including a Q&A session they had with UL, and I thought it would be interesting to pull out some of the highlights. (All material in quotes in this post comes also comes from the EE Times article, which can be found here.)

• UL has been in the security space for more than two deca
  des, and have been dealing with cybersecurity for the past decade, so they have some experience and credibility in the area. (UL was the organization that developed FIPS 140 (The Federal Information Processing Standards used by the US Government.
• They decided to create their Cyber Assurance Program (CAP) once the saw the security problems were emerging in areas beyond traditional IT. They saw that “risks are spreading out into HVAC, automotive, lighting, factory automation and medical fields” – i.e., the IoT world. CAP was in response to this, and – more directly – to a request from the Department of Homeland Security to “develop testable security criteria, through which UL can test, validate, authenticate and certify networked devices.”
• This is the testing they do: “Software used within products – ranging from chips to components and systems. We look at existing vulnerabilities, defects and patches known to third-party vendors. We test to discover coding errors and security loopholes in software, operating systems or networks. We see how a system accesses remote devices and do software updates. We offer structured penetration testing regimen, and see if we can plug those holes. We define flaws and weaknesses and provide scientific repeatable and reproducible testing criteria.”
• Of interest: they’ll test automotive components, but not vehicles themselves. Too much of the data they’ need belongs to the automotive industry, and is not available through NIST, which is the database they use for information on vulnerabilities.

Those are some of the highlights. If you’re building an IoT application (other than a connected vehicle) and you think UL certification might be useful, check out the full article, or take a look at UL.com.

I was catching up on my blog reading, when I came across an interesting post (and infographic) from TI, a long-standing technology partner of Critical Link. In the post, from this past February, Adam Yager wrote about the trend toward using connected sensors to create energy efficiency in commercial buildings. Coming on the heels of last week’s post on pushback on the Internet of Things (IoT), especially on the consumer side, it was pretty apt to see something written about an area where the IoT (and wireless sensor networks) are being adopted – and are actually producing results that justify the investment.

Let’s take a look at the infographic that Adam included in his post. The first interesting details are that commercial buildings account for 40% of global energy consumption, and – shockingly – the average commercial building wastes 30% of the energy it uses. Given that, when combined, HVAC and lighting make up 59% of commercial building energy consumption, it certainly makes sense to give the grid a break and try becoming more energy efficient.

The payback for investing in energy efficiency is significant: decreased operating costs (energy use can be lowered by 40%) and increased employee productivity. (Okay. I’m a bit skeptical about that increase in productivity, but if we just leave it at the provable metric of lower costs, it’s still worth investing.)

Here’s how sensors are used to increase energy efficiency:

“In HVAC systems, intelligent monitoring and control can be achieved by adding stand-alone environmental sensors, which will take accurate temperature and humidity readings for each zone or room throughout a building… Rather than having one central temperature-monitoring device, adding multiple sensors throughout a building gives you more options to control heating and cooling zones based on occupancy or time of day.

“Large commercial buildings can use people-counting systems for demand control ventilation (DCV). DCV adds fresh air based on the amount of people occupying a room, rather than having the HVAC system turn on based on a preset control regardless of the number of occupants present.” (Source: TI blog)

Similarly, for lighting control, connected sensors would also use occupancy detectors, as well as daylight sensors that could harvest energy, to make better use of lighting. (Many of us have experience with this approach being deployed in conference rooms that aren’t in continual use.)

One reason why the IoT has gotten more traction commercially than with a consumer audience – at least in my opinion – is because they are more immediately useful and understandable, and provide real value.  Lowering energy costs by 40% is something we can all appreciate. Being able to check your fridge via smartphone to see if you need eggs (the “storyline” from a new Samsung smart refrigerator ad) is probably not going to get many consumers too excited!

I guess it was inevitable. After all the hype. After all those forecasts of kabillions of connected devices. After all the cool videos showing smart everythings. There seems to be a bit of a backlash forming around the Internet of Things (IoT).

There’s never been any lack of muttering about IoT vulnerabilities, with tech pundits and analysts pointing out all along that that having all these IoT devices out there, many in the hands and homes of “civilians” (i.e., non-techies), was a security nightmare in the making.  But now those concerns about security have gotten through to the consumers who were supposed to be fueling the growth of IoT. A January Accenture survey, that was cited in a blog post I saw on EE Times, found that security “has moved from being a nagging problem to a top barrier as consumers are now choosing to abandon IoT devices and services over security concerns.”

While security and privacy concerns were keeping nearly half (47%) of the 28,000 respondents from buying into IoT, the top issue (62%) was how expensive IoT devices were. Nearly one-quarter (23%) couldn’t see how IoT devices would be useful to them, while 17% found them too confusing.

Thermostats, like Nest, are pretty much the only IoT-based appliance that people welcome into their home. This makes sense. The application is obvious, and the interface is pretty straightforward. (At least to me: I’ve had one for a few years.) But consumers are apparently just saying no to smart refrigerators and dishwashers. That said, I have recently begun seeing an ad for a new Samsung fridge that you can look into (via your smartphone) to check the contents. I will say that the ad doesn’t make the point very well. A guy’s at the grocery store and calls his wife to ask her whether they need eggs. She’s sitting next to the fridge, but is too involved in whatever it is she’s doing to get up and check. So the fellow whips out his smartphone and sees for himself that they do, indeed, have eggs. I’m no ad man, but wouldn’t it make more sense to have someone check the fridge who didn’t have someone sitting at home who could easily check it for him? Maybe I’m missing something here.

Anyway, in his EE Times post, Pablo Valerio points out that there’s a difference between “smart appliances” (like the fridge you can remotely peer into) and smart functions “such as dedicated sensors that can send an alert when, for example, they need some preventive maintenance or replace a critical component.”

I’m with Pablo here. I believe that consumers will welcome appliances with smart functions into their homes. Why wait for your furnace to conk out on the coldest night of the year, when you can get a call letting you know you need a service call?

I have to confess that I remain interested and intrigued by the IoT, and believe that the Internet of Everything (which is how Cisco has named it) is pretty much the future. But I’m not surprised to see that some of the hype is beginning to die down.