Paul Madsen posted an excellent article today, “Identity, Application Models and the Internet of Things,” recommending that the prevailing application development model move back to the browser and away from native apps.  He references another excellent article by Scott Jenson, “Mobile Apps Must Die,” which holds that because we use so many native mobile apps, they are “becoming too much trouble to organize and maintain,” and that the native app model, “just can’t take advantage of new opportunities.”

Paul observed how, with the prevailing native app model, the “Internet of things would push us to have 1000s of native applications on our devices, but that would place a completely unrealistic management burden on the User.”

I agree that managing large numbers of apps is becoming very burdensome and counterproductive. Each airline I fly has its own app. Each store I frequent has its own app.  I have apps upon apps upon apps.

I propose, however, that just focusing back on browser apps doesn’t completely solve the problem, particularly with the Internet of Things.  A big problem is the narrow siloed focus of so many apps.

I recently bought a Fitbit device to track all the steps I take and stairs I climb.  It is a nice little device that syncs automatically with an app on my iPhone.  I can also use that app to record food I eat and water I drink along with the automatic recording of steps and stairs.  

However, the app covers only a fairly narrow silo of functionality.  If I want to record other vital statistics (e.g blood pressure or blood glucose), it takes another app.  If I want to record my workout at the gym with any degree of granularity, it takes another app.  Of course, every app has a different concept of my identity. Not good.

Paul’s discussion of a an app to monitor his toaster begs the question – why should I have an app (either web or otherwise) for every device in my house?  Doesn’t it make more sense to have a “home management” app that accommodates toasters, fridges, thermostats, smoke alarms or whatever other Internet connected things may be available?

I propose that we need a new app paradigm that retains the great user interface characteristics of native apps, the “just in time” model of discovery and use that Paul and Scott recommend, coupled with a more integrated approach to solving real life, but more complex use cases.

 

I really enjoyed the post  Rohan Pinto tweeted about this morning – Scott Morrison’s “ We can’t let the Internet of Things become the Tyranny of Things.” Scott stated:

My belief is that the Internet of Things (IoT) will succeed or fail based on its capacity for creating its own economy. But counting devices and multiplying by people isn’t quite the right math to satisfy this equation. The real key to IoT success is how open – and more significantly, how accessible – the technology is to independent innovators.

I liked Scott’s examples of bad IoT design …

seemingly every year some earnest manufacturer actually demonstrates yet another realization of this dubious vision, which usually consists of little more than a screen stuck onto the door like some giant fridge magnet. This is IoT designed by a committee …

When I purchased my last TV, I also bought the same manufacturer’s BluRay player in the hope I could get away with one remote and hiding the latter in a closed cabinet. Boy, was I naïve.

… but his good examples were also instructive:

Take a walk into the living room and you will find an excellent example of IoT meeting its potential. IoT done right is the netfNest. A brilliant team of ex-Apple employees found a completely moribund corner of everyday technology and transformed it. They created an irresistible object of desire that quietly adapted a ponderous machine of steel and natural gas into an Internet connected device. It’s brilliant. … 

IoT done right is twiNetflix, an innovator that came up with an open API that allowed all manner of devices to integrate using simple web-based protocols. Netflix could have easily screwed this one up. They might have decided to design arcane, binary protocols optimized to support minimalist devices. Instead, they opted for open and well-documented APIs that leverage existing web understanding. The effect was to make integration accessible instead of intimidating – and in doing so, Netflix tapped into a vast developer population. The result was a Cambrian explosion of applications and devices streaming the service. You would be hard pressed to find a modern TV, disk player, or media streamer that doesn’t now have a Netflix logo somewhere on the box.

Yep, I have a plethora of ways to connect to Netflix at my house. I haven’t yet invested in the four Next thermostats I would need to control the four AC zones in my house, but two of my sons have them.

In closing, Scott challenges us:

It’s time to worry less about trying to make the Internet of Things something different. Instead, we need to focus on making it more of the same, more like, well, the internet. Declare IoT open, base it on APIs, and then step back and watch the engine of Silicon Valley engage.

Well spoken, Scott. And thanks for introducing me to that eminently hashtaggable acronym: #IoT.

Some interesting ideas are swirling in my mind in response to Ian Glazer’s challenge, “Killing IAM in Order to Save It” and Dave Kearn’s article “Pervasive and Ubiquitous Identity.”

Whether or not we need, as Ian suggests, to completely restructure IAM systems in order to progress is still subject for debate, but the concept of thinking about and representing relationships between identities in a directed graph format is intriguing to me.

According to Wikipedia, “Graph databases are based on graph theory. Graph databases employ nodes, properties, and edges.” The following diagram gives a simple example. 

 

Using this method, we can visualize identities as nodes, each with relevant properties, and relationships between identities as edges.  Interestingly, the edges, or relationships, may also have identities and properties of their own.  

As Dave suggests, identities are not only for people, but for things, platforms and services.  The simple diagram below begins to illustrate this concept:

 

 

 

The relationships (edges) are primarily verbs that describe what actions the relationship supports.  A primary role of identity management systems is to establish these relationships between people identities and service or thing identities in such a way that valuable actions can be performed.

These are a few of my thoughts.  What do you think?

PS. Can anyone recommend a good directed-graph drawing tool for Mac?

I attended a very thought-provoking Kuppinger Cole webinar last week, entitled, “SAML is Dead.  Long Live SAML,” featuring Craig Burton of Kuppinger Cole and Pam Dingle of Ping Identity.  It is now available as an on demand webcast.  My favorite slide addressed the sheer scale of what we are expecting to see in just a few years.

We are all familiar with big, complex operations now:

• Large enterprise Identity repositories:  hundreds of thousands
• Large mobile telephony user repositories: low hundreds of millions
• Large social media sites: high hundreds of millions

Adding addressable devices and the API’s to support those devices is mind boggling.

• Devices by 2015:  almost 3 billion
• API’s to support all those devices: almost 27 billion

Meeting that demand will take some real innovative technology and processes.  The webcast was certainly worth an hour of my time.  I highly recommend it to you.

Having a deep fascination with The Internet of Things, I thoroughly enjoyed reading Phil Windley’s recent post, “Personal Event Networks: Building the Internet of Things,” and Drummond Reed’s commentary, “Phil Windley on Personal Event Networks.”

Phil concludes in his post, “An Internet of Things—social products and services—will have as profound an effect on our lives as the changes of the preceding 15 years. I believe that personalized event-driven programming models are a key part of the architecture that makes them real.”

In his post, Drummond states, “Many things become possible if your personal network of devices, products, and services can safely talk to each other in ways they can all understand. That’s what Phil is promoting through a simple event interface.”

It occurred to me that Identity is a key enabler – the Identities of People meeting the Identities of Things.  What transpires will be meaningful relationships between people and the things which provide services to them.  I like to think I already have meaningful relationships with things like my refrigerator and my car (I’m weird that way), but think such relationships can be significantly enhanced as the Internet of Things evolves.

I applaud the pioneering work of Drummond and Phil and others like them, who are working to bring about meaningful reality to these fascinating concepts.

Last month, I created a series of posts (one, two, three) about the Internet of Things.  I turns out that one of my colleagues who inspired that series of blog posts is now employed by Tendril, a company that is involved in this Internet of Things business for real.

According to their press release footer:

Tendril is a leading energy platform company that is helping to drive the large-scale deployment of the Smart Grid through the development of forward-thinking solutions as well as its work to establish industry protocols. The Tendril platform provides an open standards-based, scalable and secure end-to-end solution for the Energy Internet – the network for existing and upcoming Smart Grid technologies. With applications, products and services enabled by the platform, Tendril creates a dialogue and marketplace between energy providers, consumers and the energy ecosystem.

The Smart Grid concept will certainly be involved in attaching lots of devices to the Internet. For one intriguing project, Tendril has teamed with Whirlpool to focus on the roll-out of smart home appliances in the US. For example,

For a refrigerator to actively manage its energy consumption, it must be able to quickly, reliably and seamlessly communicate with the electric utility company. … In this case, the refrigerator will automatically move its defrost cycle to a non-peak time without impacting the performance of the appliance.

I like the idea of having smart appliances coordinate with the electric utility to save energy and reduce my energy bill. It will be great to see what companies like Tendril will do to productively contribute to the Internet of Things.

The concept of the Internet of Things is my  fascination de jour.  Today, I was pleased to read an intriguing Fast Company article entitled, “Wi-Fi-Connected Lightbulbs, Coming To Smart Homes In 2012.”   It appears that WiFi enabled home lighting is closer that I thought, from Netherlands-based company NXP:

Why on Earth would you want a lightbulb with an IP address? It’s not obvious until you realize we’re not talking regular incandescent bulbs here. The tech will go into advanced compact fluorescent units as well as LED light bulbs, both clean low-power replacements for Edison’s aging invention. These lights already incorporate a few chunks of silicon in their bases to help control them, and it’s this tiny circuit board that enables all sorts of new things–adding NXP’s tiny Wi-Fi system to the board is relatively easy and cheap. And then you can turn your lights on and off from a computer hooked up to your home’s wireless grid.

I’ve always thought it would be cool to control all the lights in my house from a single point.  The fact that next year I could do that without rewiring my house  – and do the controlling from my mobile phone or other Internet-connected device - is pretty cool!

Craig Burton offered an intriguing comment on my recent blog post about the Live Web and Interent of Things.  He referred me to two projects that used the Arduino open source electronics prototyping platform, Garduino, a computer that runs your garden and Yobot, the Arduino-based Yogurt Maker!

As pointed out by Craig in his comment, we could add WiFi or Ethernet connections to these projects and they would become part of the Internet of Things, connected in a loosely coupled way.

The Arduino platform could be used to prototype all sorts of interesting “Internet of Things” applications.

This reminded me that way back in 2008, I blogged about efforts by my Sun Microsystem colleagues Louis Pfortmiller and Guanghwa Ho to integrate small devices into the Sun Project Destination architecture for highly-personalized online services.  Their efforts involved the use of “SunSPOT” wireless sensor/actuator devices that were programmed in Java.  We discussed all sorts of intriguing uses for this technology It turns out that this project apparently still exists in Oracle Labs:

This almost makes me want to resurrect my old electronics hobby and dive into adding things to the Internet.

Craig Burton, who recently joined Kuppinger Cole, authored an insightful article, “Bringing the Web to Life at Last,” addressing two compelling topics: “The Live Web” and the “Internet of Things“.  His final statement provides an apt summary:

We don’t need a Facebook of Things. We need an Internet of Things. We need the Live Web.

Craig outlines the basis concepts of the Live Web:

The term Live Web was first coined by Doc Searls and his son Allen Searls to describe a Web where timeliness and context matter as much as relevance. It blossoms with the following three assumptions:

• All things are connected to the Internet.
• All things are recorded and tagged.
• All things can be recalled and accessed in context.

The Live Web is made up of three core principles that give rise to generating context:

• First principle: Ubiquitous programmable data access. (APIs)
• Second principle: Ubiquitous event-based endpoints.
• Third principle: Ubiquitous event-based evaluation and execution machines.

Note that the three principles match the three assumptions.

Craig explains how, even if all devices are connected to the Internet (the Internet of Things), the current web paradigm, as wonderful as it is, would not work well, because the current web operates as a tightly-coupled manner, like Facebook:

The problem with the idea of a big “Facebook of Things” kind of site is the tight coupling that it implies a person would have to take charge of all the devices. You would have to “friend” each one. And remember, these are devices, so not only do you have to connect and “friend” them but you will be doing the work of managing them.

…

This just isn’t going to happen. Ever.

However, applying Live Web principles and loosely coupling the device will enable the Internet of Things to work:

Each device can interpret that message however it sees fit or ignore it altogether. This significantly reduces the complexity of the overall system because individual devices are loosely coupled.

It will be fun to see the progressive realization of these concepts: all devices connected to the Internet and coupled in a loose sort of way that makes possible all kinds of interesting applications.  I just  wonder how long will it be before I can sit here at my desk and command my Live Web connected kitchen to cook me up a nice omelette for breakfast?