Showing posts with label Internet of things. Show all posts
Showing posts with label Internet of things. Show all posts

Thursday, March 19, 2015

Axeda Builds Machine Cloud on HP Vertica to Deliver On-Demand Analysis Services

Transcript of a BriefingsDirect discussion on how to manage machine-to-machine communication for better big data collection and analysis.

Listen to the podcast. Find it on iTunes. Download the transcript. Sponsor: HP.

Dana Gardner: Hello, and welcome to the next edition of the HP Discover Podcast Series. I'm Dana Gardner, Principal Analyst at Interarbor Solutions, your host and moderator for this ongoing sponsored discussion on IT innovation and how it’s making an impact on people’s lives. Once again, we're focusing on how companies are adapting to the new style of IT to improve IT performance and deliver better user experiences, as well as better business results.
Gardner

Our next innovation case study interview highlights how Axeda, based in Foxboro,  Massachusetts, is creating a machine-to-machine (M2M) capability for analysis -- in other words, an Axeda Machine Cloud.

We're going to learn how they partner with HP in doing that. We're joined in our discussion today by Kevin Holbrook, the Senior Director of Advance Development at Axeda. Welcome, Kevin.

Kevin Holbrook: Thank you for having me.

Gardner: We have the whole Internet of Things (IoT) phenomenon. People are accepting more and more devices, end points, sensors, even things within the human body, delivering data out to applications and data pools. What do you do in terms of helping organizations start to come to grip with this M2M and Internet of Things data demand?
Become a member of myVertica
Register now
Gain access to the HP Vertica Community Edition
Holbrook: It starts with the connectivity space. Our focus has largely been in OEMs, equipment manufacturers. These are people who have the "M" in the M2M or the "T" in the Internet of Things. They are manufacturing things.

The initial drivers to have a handle on those things are basic questions, such as, "Is this device on?" There are multi-million dollar machines that are currently deployed in the world where that question can’t be answered without a phone call.

Initial driver

That was the initial driver, the seed, if you will. We entered into that space from the remote-service angle. We deployed small-agent software to the edge to get the first measurements from those systems and get them pushed up to the cloud, so that users can interact with it.

Holbrook
That grew into remote accesstelnet sessions or remote desktop being able to physically get down there, debug, tweak, and look at the devices that are operating. From there, we grew into software distribution, or content distribution. That could be anything from firmware updates to physically distributing configuration and calibration files for the instrument. We're recently seeing an uptake in content distribution for things like digital signage or in-situ ads being displayed on consumer goods.

From there, we started aggregating data. We have about 1.5 million assets connected to our cloud now globally, and there is all kinds of data coming in. Some of it's very, very basic from a resource standpoint, looking at CPU consumption, disks space, available memory, things of that nature.

It goes all the way through to usage and diagnostics, so that you can get a very granular impression how this machine is operating. As you begin to aggregate this data, all sorts of challenges come out of it. HP has proven to be a great partner for starting to extract value.

We can certainly get to the data, we can connect the device, and we can aggregate that data to our partners or to the customer directly. Getting value from that data is a completely different proposition. Data for data’s sake is not high value.
From our perspective, Vertica represents an endpoint. We've carried the data, cared for the data, and made sure that the device was online, generating the right information and getting it into Vertica.

Gardner:  What is it that you're using Vertica for to do that? Are we creating applications, are we giving analysis as a service? How is this going to market for you?

Holbrook: From our perspective, Vertica represents an endpoint. We've carried the data, cared for the data, and made sure that the device was online, generating the right information and getting it into Vertica.

When we approach customers, were approaching it from a joint-sale perspective. We're the connectivity layer, the instrumentation, the business automation layer there, and we're getting it into Vertica ,so that can be the seed for applications for business intelligence (BI) and for analytics.

So, we are the lowest component in the stack when we walk into one of these engagements with Vertica. Then, it's up to them, on a customer-by-customer basis, to determine what applications to bring to the table. A lot of that is defined by the group within the organization that actually manages connectivity.

We find that there's a big difference between a service organization, which is focused primarily on keeping things up and running, versus a business unit that’s driving utilization metrics, trying to determine not only how things are used, but how it can influence their billing.

Business use

We've found that that's a place where Vertica has actually been quite a pop for us in talking to customers. They want to know not just the simple metrics of the machines' operation, but how that reflects the business use of it.

The entire market has shifted and continues to shift. I was somewhat taken aback only a couple of weeks ago, when I found out that you can no longer buy a jet engine. I thought this was a piece of hardware you purchased, as opposed to something that you may have rented and paid per use. And so [the model changes to leasing] as the machines get  bigger and bigger. We have GE and the Bureau of Engraving and Printing as customers.

We certainly have some very large machines connected to our cloud and we're finding that these folks are shifting away from the notion that one owns a machine and consumes it until it breaks or dies. Instead, one engages in an ongoing service model, in which you're paying for the use of that machine.

While we can generate that data and provide some degree of visibility and insight into that data, it takes a massive analytics platform to really get the granular patterns that would drive business decisions.

Gardner: It sounds like many of your customers have used this for some basic blocking and tackling about inventory and access and control, then moved up to a business metrics of how is it being used, how we're billing, audit trails, and that sort of thing. Now, we're starting to look at a whole new type of economy. It's a services economy, based on cloud interactivity, where we can give granular insights, and they can manage their business very, very tightly.
There's not only a ton of data being generated, but the regulatory and compliance requirements which dictate where you can even leave that data at rest.

Any thoughts about what's going to be required of your organization to maintain scale? The more use cases and the more success, of course, the more demand for larger data and even better analytics. How do you make sure that you don't run out of runway on this?

Holbrook: There are a couple of strategies we've taken, but before I dive into that, I'll say that the issue is further complicated by the issue of data homing. There's not only a ton of data being generated, but the regulatory and compliance requirements which dictate where you can even leave that data at rest. Just moving it around is one problem, and where it sits on a disk is a totally different problem. So we're trying to tackle all of these.

The first way to address the scale for us from an architectural perspective was to try to distribute the connectivity. In order for you to know that something's running, you need to hear from it. You might be able to reach out, what we call contactability, to say, "Tell me if you're still running." But, by and large, you know of a machine's existence and its operation by virtue of it telling you something. So even if a message is nothing more than "Hello, I'm here," you need to hear from this device.

From the connectivity standpoint, our goal is not to try to funnel all of this into a single pipe, but rather to find where to get a point of presence that is closest and that is reasonable. We’ve been doing this on our remote-access technology for years, trying to find the appropriate geographically distributed location to route data through, to provide as easy and seamless an experience as possible.

So that’s the first, as opposed to just ruthlessly federating all incoming data, distributing the connectivity infrastructure, as well as trying to get that data routed to its end consumer as quickly as possible.

We break down data from our perspective into three basic temporal categories. There's the current data, which is the value you would see reading a dial on the machine. There's recent data, which would tell you whether something is trending in a negative direction, say pressure going up. Then, there is the longer-term historical data. While we focus on the first two, we’d deliberately, to handle the scale problem, don't focus on the long-term historical data.

Recent data

I'll treat recent data as being anywhere from 7 to 120 days and beyond, depending on the data aggregation rates. We focus primarily on that. When you start to scale beyond that, where the real long tail of this is, we try to make sure that we have our partner in place to receive the data.

We don't want to be diving into two years of data to determine seasonal trending when we're attempting to collect data from 1.5 million assets and acting as quickly as possible to respond to error conditions at the edge.

Gardner: Kevin, what about the issue of latency? I imagine some of your customers have a very dire need to get analysis very rapidly on an ongoing streamed basis. Others might be more willing to wait and do it in a batch approach in terms of their analytics. How do you manage that, and what are some of the speeds and feeds about the best latency outcomes?

Holbrook: That’s a fantastic question. Everybody comes in and says we need a zero-latency solution. Of course, it took them about two-and-a-half seconds to say that.
Become a member of myVertica
Register now
Gain access to the HP Vertica Community Edition
There's no such thing as real-time, certainly on the Internet. Just negotiating up the TCP stack and tearing it down to send one byte is going to take you time. Then, we send it over wires under the ocean, bounce it off a satellite, you name it. That's going to take time.

There are two components to it. One is accepting that near-real-time, which is effectively the transport latency, is the smallest amount of time it can take to physically go from point A to point B, absent having a dedicated fiber line from one location to the other. We can assume that on the Internet that's domestically somewhere in the one- to two-second range. Internationally, it's in the two- to three-second or beyond range, depending on the connectivity of the destination.

What we provide is an ability to produce real-time streams of data outbound. You could take from one asset, break up the information it generates, and stream it to multiple consumers in near-real-time in order to get the dashboard in the control center to properly reflect the state of the business. Or you can push it to a data warehouse in the back end, where it then can be chunked and ETLd into some other analytics tool.

For us, we try not to do the batch ETLing. We'd rather make sure that we handle what we're good at. We're fantastic at remote service, at automating responses, at connectivity and at expanding what we do. But we're never going to be a massive ETL, transforming and converting into somebody’s data model or trying to get deep analytics as a result of that.

Gardner: Was it part of this need for latency, familiarity, and agility that led into Vertica? What were some of the decisions that led to picking Vertica as a partner?

Several reasons

Holbrook: There were a few reasons. That was one of them. Also the fact that there's a massive set of offerings already on top of it. A lot of the other people when we considered this -- and I won't mention competitors that we looked at -- were more just a piece of the stack, as opposed to a place where solutions grew out of.

It wasn't just Vertica, but the ecosystem built on top of Vertica. Some of the vendors we looked at are currently in the partner zone, because they're now building their solutions on top of Vertica.

We looked at it as an entry point into an ecosystem and certainly the in-memory component, the fact that you're getting no disk reads for massive datasets was very attractive for us. We don’t want to go through that process. We've dealt with the struggles internally of trying to have a relational data model scale. That’s something that Vertica has absolutely solved.

Gardner: Now your platform includes application services, integration framework, and data management. Let’s hone in on the application services. How are developers interested in getting access to this? What are their demands in terms of being able to use analysis outcomes, outputs, and then bring that into an application environment that they need to fulfill their requirements to their users?
It wasn't just Vertica, but the ecosystem built on top of Vertica. Some of the vendors we looked at are currently in the partner zone, because they're now building their solutions on top of Vertica.

Holbrook: It breaks them down into two basic categories. The first is the aggregation and the collection of data, and the second is physical interaction with the device. So we focus on both about equally. When we look at what developers are doing, almost always it’s transforming the data coming in and reaching out to things like a customer relationship management (CRM) system. It's opening a ticket when a device has thrown a certain error code or integrating with a backend drop-ship distribution system in the event that some consumable has begun to run low.

In terms of interaction, it's been significant. On the data side, we primarily see that they're  extracting subsets of data for deeper analysis. Sometimes, this comes up in discrete data points. Frequently, this comes up in the transfer of files. So there is a certain granularity that you can survive. Coming down the fire-hose is discrete data points that you can react to, and there's a whole other order of magnitude of data that you can handle when it's shipped up in a bulk chunk.

A good example is one of the use cases we have with GE in their oil and gas division  where they have a certain flow of data that's always ongoing and giving key performance indicators (KPIs). But this is nowhere near the level of data that they're actually collecting. They have database servers that are co-resident with these massive gas pipeline generators.

So we provide them the vehicle for that granular data. Then, when a problem is detected automatically, they can say, "Give me far more granular data for the problem area." it could be five minutes before or five minutes since. This is then uploaded, and we hand off to somewhere else.

So when we find developers doing integration around the data in particular, it's usually when they're diving in more deeply based on some sort of threshold or trigger that has been encountered in the field.

Gardner: And lastly, Kevin, for other organizations that are looking to create data services and something like your Axeda Machine Cloud, are there any lessons learned that you could share when it comes to managing such complexity, scale, and the need for speed? What have you learned at a high level that you could share?

All about strategy

Holbrook: It’s all going to be about the data-collection strategy. You're going to walk into a customer or potential customer, and their default response is going to be, "Collect everything." That’s not inherently valuable. Just because you've collected it, doesn’t mean that you are going to get value from it. We find that, oftentimes, 90-95 percent of the data collected in the initial deployment is not used in any constructive way.

I would say focus on the data collection strategy. Scale of bad data is scale for scale’s sake. It doesn’t drive business value. Make sure that the folks who are actually going to be doing the analytics are in the room when you are doing your data collection strategy definition. when you're talking to the folks who are going to wire up sensors,  and when you're talking to the folks who are building the device.

Unfortunately, these are frequently within a larger business ,in particular, completely different groups of people that might report to completely different vice presidents. So you go to one group, and they have the connectivity guys. You talk about it and you wire everything up.
We find that, oftentimes, 90-95 percent of the data collected in the initial deployment is not used in any constructive way.

Then, six to eight months later, you walk into another room. They’ll say "What the heck is this? I can’t do anything with this. All I ever needed to know was the following metric." It wasn’t collected because the two hadn't stayed in touch. The success of deployed solutions and the reaction to scale challenges is going to be driven directly by that data-collection strategy. Invest the time upfront and then you'll have a much better experience in the back.

Gardner: Very good. I'm afraid we’ll have to leave it there. We've been learning how Axeda in Foxboro Massachusetts is providing services to its customers to the Axeda Machine Cloud and as a partner with HP as using their Vertica Analysis Platform to provide those insights.
Become a member of myVertica
Register now
Gain access to the HP Vertica Community Edition
So I'd like to thank our guest. We've been joined by Kevin Holbrook, the Senior Director of Advanced Development at Axeda. Thank you, Kevin.

Holbrook: Thank you.

Gardner: And a big thank you to our audience for joining us for the special new style of IT discussion. I'm Dana Gardner; Principal Analyst at Interarbor Solutions, your host for this ongoing series of HP sponsored discussions. Thanks again for listening, and come back next time.

Listen to the podcast. Find it on iTunes. Download the transcript. Sponsor: HP.

Transcript of a BriefingsDirect discussion on how to manage machine-to-machine communication for better big data collection and analysis. Copyright Interarbor Solutions, LLC, 2005-2015. All rights reserved.

You may also be interested in:

Monday, September 22, 2014

The Open Group Panel: Internet of Things Poses Opportunities and Obstacles

Transcript of The Open Group podcast, in conjunction with BriefingsDirect, exploring the challenges and ramifications of the Internet of Things, as machines and sensors collect vast amounts of data.

Listen to the podcast. Find it on iTunes. Download the transcript. Sponsor: The Open Group.

Dana Gardner: Hello, and welcome to a special BriefingsDirect thought leadership interview series coming to you in conjunction with recent The Open Group Boston 2014 on July 21 in Boston.

Gardner
I'm Dana Gardner, principal analyst at Interarbor Solutions, and I'll be your host and moderator throughout these discussions on Open Platform 3.0 and Boundaryless Information Flow.

We're going to now specifically delve into the Internet of Things with a panel of experts. The conference has examined how Open Platform 3.0 leverages the combined impacts of cloud, big data, mobile, and social. But to each of these now we can add a new cresting wave of complexity and scale as we consider the rapid explosion of new devices, sensors, and myriad endpoints that will be connected using internet protocols, standards and architectural frameworks.

This means more data, more cloud connectivity and management, and an additional tier of “things” that are going to be part of the mobile edge -- and extending that mobile edge ever deeper into even our own bodies.

When we think about inputs to these social networks -- that's going to increase as well. Not only will people be tweeting, your device could be very well tweet, too -- using social networks to communicate. Perhaps your toaster will soon be sending you a tweet about your English muffins being ready each morning.

The Internet of Things is more than the “things” – it means a higher order of software platforms. For example, if we are going to operate data centers with new dexterity thanks to software-defined networking (SDN) and storage (SDS) -- indeed the entire data center being software-defined (SDDC) -- then why not a software-defined automobile, or factory floor, or hospital operating room -- or even a software-defined city block or neighborhood?

And so how does this all actually work? Does it easily spin out of control? Or does it remain under proper management and governance? Do we have unknown unknowns about what to expect with this new level of complexity, scale, and volume of input devices?

Will architectures arise that support the numbers involved, interoperability, and provide governance for the Internet of Things -- rather than just letting each type of device do its own thing?

To help answer some of these questions, The Open Group assembled a distinguished panel to explore the practical implications and limits of the Internet of Things. So please join me in welcoming Said Tabet, Chief Technology Officer for Governance, Risk and Compliance Strategy at EMC, and a primary representative to the Industrial Internet Consortium; Penelope Gordon, Emerging Technology Strategist at 1Plug Corporation; Jean-Francois Barsoum, Senior Managing Consultant for Smarter Cities, Water and Transportation at IBM, and Dave Lounsbury, Chief Technical Officer at The Open Group.

Jean-Francois, we have heard about this notion of "cities as platforms," and I think the public sector might offer us some opportunity to look at what is going to happen with the Internet of Things, and then extrapolate from that to understand what might happen in the private sector.

Hypothetically, the public sector has a lot to gain. It doesn't have to go through the same confines of a commercial market development, profit motive, and that sort of thing. Tell us a little bit about what the opportunity is in the public sector for smart cities.

Jean-Francois Barsoum: It's immense. The first thing I want to do is link to something that Marshall Van Alstyne (Professor at Boston University and Researcher at MIT) had talked about, because I was thinking about his way of approaching platforms and thinking about how cities represent an example of that.

Barsoum
You don't have customers; you have citizens. Cities are starting to see themselves as platforms, as ways to communicate with their customers, their citizens, to get information from them and to communicate back to them. But the complexity with cities is that as a good a platform as they could be, they're relatively rigid. They're legislated into existence and what they're responsible for is written into law. It's not really a market.

Chris Harding (Forum Director of The Open Group Open Platform 3.0) earlier mentioned, for example, water and traffic management. Cities could benefit greatly by managing traffic a lot better.

Part of the issue is that you might have a state or provincial government that looks after highways. You might have the central part of the city that looks after arterial networks. You might have a borough that would look after residential streets, and these different platforms end up not talking to each other.

They gather their own data. They put in their own widgets to collect information that concerns them, but do not necessarily share with their neighbor. One of the conditions that Marshall said would favor the emergence of a platform had to do with how much overlap there would be in your constituents and your customers. In this case, there's perfect overlap. It's the same citizen, but they have to carry an Android and an iPhone, despite the fact it is not the best way of dealing with the situation.

The complexities are proportional to the amount of benefit you could get if you could solve them.

Gardner: So more interoperability issues?

Barsoum: Yes.

More hurdles

Gardner: More hurdles, and when you say commensurate, you're saying that the opportunity is huge, but the hurdles are huge and we're not quite sure how this is going to unfold.

Barsoum: That's right.

Gardner: Let's go to an area where the opportunity outstrips the challenge, manufacturing. Said, what is the opportunity for the software-defined factory floor for recognizing huge efficiencies and applying algorithmic benefits to how management occurs across domains of supply-chain, distribution, and logistics. It seems to me that this is a no-brainer. It's such an opportunity that the solution must be found.

Tabet: When it comes to manufacturing, the opportunities are probably much bigger. It's where we can see a lot of progress that has already been done and still work is going on. There are two ways to look at it.

Tabet
One is the internal side of it, where you have improvements of business processes. For example, similar to what Jean-Francois said, in a lot of the larger companies that have factories all around the world, you'll see such improvements on a factory base level. You still have those silos at that level.

Now with this new technology, with this connectedness, those improvements are going to be made across factories, and there's a learning aspect to it in terms of trying to manage that data. In fact, they do a better job. We still have to deal with interoperability, of course, and additional issues that could be jurisdictional, etc.

However, there is that learning that allows them to improve their processes across factories. Maintenance is one of them, as well as creating new products, and connecting better with their customers. We can see a lot of examples in the marketplace. I won't mention names, but there are lots of them out there with the large manufacturers.

Gardner: We've had just-in-time manufacturing and lean processes for quite some time, trying to compress the supply chain and distribution networks, but these haven't necessarily been done through public networks, the internet, or standardized approaches.

But if we're to benefit, we're going to need to be able to be platform companies, not just product companies. How do you go from being a proprietary set of manufacturing protocols and approaches to this wider, standardized interoperability architecture?

Tabet: That's a very good question, because now we're talking about that connection to the customer. With the airline and the jet engine manufacturer, for example, when the plane lands and there has been some monitoring of the activity during the whole flight, at that moment, they'll get that data made available. There could be improvements and maybe solutions available as soon as the plane lands.

Interoperability

That requires interoperability. It requires Platform 3.0 for example. If you don't have open platforms, then you'll deal with the same hurdles in terms of proprietary technologies and integration in a silo-based manner.

Gardner: Penelope, you've been writing about the obstacles to decision-making that might become apparent as big data becomes more prolific and people try to capture all the data about all the processes and analyze it. That's a little bit of a departure from the way we've made decisions in organizations, public and private, in the past.

Of course, one of the bigger tenets of Internet of Things is all this great data that will be available to us from so many different points. Is there a conundrum of some sort? Is there an unknown obstacle for how we, as organizations and individuals, can deal with that data? Is this going to be chaos, or is this going to be all the promises many organizations have led us to believe around big data in the Internet of Things?

Penelope Gordon: It's something that has just been accelerated. This is not a new problem in terms of the decision-making styles not matching the inputs that are being provided into the decision-making process.

Gordon
Former US President Bill Clinton was known for delaying making decisions. He's a head-type decision-maker and so he would always want more data and more data. That just gets into a never-ending loop, because as people collect data for him, there is always more data that you can collect, particularly on the quantitative side. Whereas, if it is distilled down and presented very succinctly and then balanced with the qualitative, that allows intuition to come to fore, and you can make optimal decisions in that fashion.

Conversely, if you have someone who is a heart-type or gut-type decision-maker and you present them with a lot of data, their first response is to ignore the data. It's just too much for them to take in. Then you end up completely going with whatever you feel is correct or whatever you have that instinct that it's the correct decision. If you're talking about strategic decisions, where you're making a decision that's going to influence your direction five years down the road, that could be a very wrong decision to make, a very expensive decision, and as you said, it could be chaos.

It just brings to mind to me Dr. Seuss’s The Cat in the Hat with Thing One and Thing Two. So, as we talk about the Internet of Things, we need to keep in mind that we need to have some sort of structure that we are tying this back to and understanding what are we trying to do with these things.
If you have someone who is a heart-type or gut-type decision-maker and you present them with a lot of data, their first response is to ignore the data.

Gardner: Openness is important, and governance is essential. Then, we can start moving toward higher-order business platform benefits. But, so far, our panel has been a little bit cynical. We've heard that the opportunity and the challenges are commensurate in the public sector and that in manufacturing we're moving into a whole new area of interoperability, when we think about reaching out to customers and having a boundary that is managed between internal processes and external communications.

And we've heard that an overload of data could become a very serious problem and that we might not get benefits from big data through the Internet of Things, but perhaps even stumble and have less quality of decisions.

So Dave Lounsbury of The Open Group, will the same level of standardization work? Do we need a new type of standards approach, a different type of framework, or is this a natural path and course what we have done in the past?

Different level

Dave Lounsbury: We need to look at the problem at a different level than we institutionally think about an interoperability problem. Internet of Things is riding two very powerful waves, one of which is Moore's Law, that these sensors, actuators, and network get smaller and smaller. Now we can put Ethernet in a light switch right, a tag, or something like that.

Lounsbury
Also, Metcalfe's Law that says that the value of all this connectivity goes up with the square of the number of connected points, and that applies to both the connection of the things but more importantly the connection of the data.

The trouble is, as we have said, that there's so much data here. The question is how do you manage it and how do you keep control over it so that you actually get business value from it. That's going to require us to have this new concept of a platform to not only to aggregate, but to just connect the data, aggregate it, correlate it as you said, and present it in ways that people can make decisions however they want.

Also, because of the raw volume, we have to start thinking about machine agency. We have to think about the system actually making the routine decisions or giving advice to the humans who are actually doing it. Those are important parts of the solution beyond just a simple "How do we connect all the stuff together?"

Gardner: We might need a higher order of intelligence, now that we have reached this border of what we can do with our conventional approaches to data, information, and process.

Thinking about where this works best first in order to then understand where it might end up later, I was intrigued again this morning by Professor Van Alstyne. He mentioned that in healthcare, we should expect major battles, that there is a turf element to this, that the organization, entity or even commercial corporation that controls and manages certain types of information and access to that information might have some very serious platform benefits.
The question is how do you manage it and how do you keep control over it so that you actually get business value from it.

The openness element now is something to look at, and I'll come back to the public sector. Is there a degree of openness that we could legislate or regulate to require enough control to prevent the next generation of lock-in, which might not be to a platform to access to data information and endpoints? Where is it in the public sector that we might look to a leadership position to establish needed openness and not just interoperability.

Barsoum: I'm not even sure where to start answering that question. To take healthcare as an example, I certainly didn't write the bible on healthcare IT systems and if someone did write that, I think they really need to publish it quickly.

We have a single-payer system in Canada, and you would think that would be relatively easy to manage. There is one entity that manages paying the doctors, and everybody gets covered the same way. Therefore, the data should be easily shared among all the players and it should be easy for you to go from your doctor, to your oncologist, to whomever, and maybe to your pharmacy, so that everybody has access to this same information.

We don't have that and we're nowhere near having that. If I look to other areas in the public sector, areas where we're beginning to solve the problem are ones where we face a crisis, and so we need to address that crisis rapidly.

Possibility of improvement

In the transportation infrastructure, we're getting to that point where the infrastructure we have just doesn't meet the needs. There's a constraint in terms of money, and we can't put much more money into the structure. Then, there are new technologies that are coming in. Chris had talked about driverless cars earlier. They're essentially throwing a wrench into the works or may be offering the possibility of improvement.

On any given piece of infrastructure, you could fit twice as many driverless cars as cars with human drivers in them. Given that set of circumstances, the governments are going to find they have no choice but to share data in order to be able to manage those. Are there cases where we could go ahead of a crisis in order to manage it? I certainly hope so.

Gardner: How about allowing some of the natural forces of marketplaces, behavior, groups, maybe even chaos theory, where if sufficient openness is maintained there will be some kind of a pattern that will emerge? We need to let this go through its paces, but if we have artificial barriers, that might be thwarted or power could go to places that we would regret later.

Barsoum: I agree. People often focus on structure. So the governance doesn't work. We should find some way to change the governance of transportation. London has done a very good job of that. They've created something called Transport for London that manages everything related to transportation. It doesn't matter if it's taxis, bicycles, pedestrians, boats, cargo trains, or whatever, they manage it.
In the transportation infrastructure, we're getting to that point where the infrastructure we have just doesn't meet the needs.

You could do that, but it requires a lot of political effort. The other way to go about doing it is saying, "I'm not going to mess with the structures. I'm just going to require you to open and share all your data." So, you're creating a new environment where the governance, the structures, don't really matter so much anymore. Everybody shares the same data.

Gardner: Said, to the private sector example of manufacturing, you still want to have a global fabric of manufacturing capabilities. This is requiring many partners to work in concert, but with a vast new amount of data and new potential for efficiency.

How do you expect that openness will emerge in the manufacturing sector? How will interoperability play when you don't have to wait for legislation, but you do need to have cooperation and openness nonetheless?

Tabet: It comes back to the question you asked Dave about standards. I'll just give you some examples. For example, in the automotive industry, there have been some activities in Europe around specific standards for communication.

The Europeans came to the US and started to have discussions, and the Japanese have interest, as well as the Chinese. That shows, because there is a common interest in creating these new models from a business standpoint, that these challenges they have to be dealt with together.

Managing complexity

When we talk about the amounts of data, what we call now big data, and what we are going to see in about five years or so, you can't even imagine. How do we manage that complexity, which is multidimensional? We talked about this sort of platform and then further, that capability and the data that will be there. From that point of view, openness is the only way to go.

There's no way that we can stay away from it and still be able to work in silos in that new environment. There are lots of things that we take for granted today. I invite some of you to go back and read articles from 10 years ago that try to predict the future in technology in the 21st century. Look at your smart phones. Adoption is there, because the business models are there, and we can see that progress moving forward.

Collaboration is a must, because it is a multidimensional level. It's not just manufacturing like jet engines, car manufacturers, or agriculture, where you have very specific areas. They really they have to work with their customers and the customers of their customers.
Adoption is there, because the business models are there, and we can see that progress moving forward.

Gardner: Dave, I have a question for both you and Penelope. I've seen some instances where there has been a cooperative endeavor for accessing data, but then making it available as a service, whether it's an API, a data set, access to a data library, or even analytics applications set. The Ocean Observatories Initiative is one example, where it has created a sensor network across the oceans and have created data that then they make available.

Do you think we expect to see an intermediary organization level that gets between the sensors and the consumers or even controllers of the processes? Is there's a model inherent in that that we might look to -- something like that cooperative data structure that in some ways creates structure and governance, but also allows for freedom? It's sort of an entity that we don't have yet in many organizations or many ecosystems and that needs to evolve.

Lounsbury: We're already seeing that in the marketplace. If you look at the commercial and social Internet of Things area, we're starting to see intermediaries or brokers cropping up that will connect the silo of my android ecosystem to the ecosystem of package tracking or something like that. There are dozens and dozens of these cropping up.

In fact, you now see APIs even into a silo of what you might consider a proprietary system and what people are doing is to to build a layer on top of those APIs that intermediate the data.

This is happening on a point-to-point basis now, but you can easily see the path forward. That's going to expand to large amounts of data that people will share through a third party. I can see this being a whole new emerging market much as what Google did for search. You could see that happening for the Internet of Things.

Gardner: Penelope, do you have any thoughts about how that would work? Is there a mutually assured benefit that would allow people to want to participate and cooperate with that third entity? Should they have governance and rules about good practices, best practices for that intermediary organization? Any thoughts about how data can be managed in this sort of hierarchical model?

Nothing new

Gordon: First, I'll contradict it a little bit. To me, a lot of this is nothing new, particularly coming from a marketing strategy perspective, with business intelligence (BI). Having various types of intermediaries, who are not only collecting the data, but then doing what we call data hygiene, synthesis, and even correlation of the data has been around for a long time.

It was an interesting, when I looked at recent listing of the big-data companies, that some notable companies were excluded from that list -- companies like Nielsen. Nielsen's been collecting data for a long time. Harte-Hanks is another one that collects a tremendous amount of information and sells that to companies.

That leads into the another part of it that I think there's going to be. We're seeing an increasing amount of opportunity that involves taking public sources of data and then providing synthesis on it. What remains to be seen is how much of the output of that is going to be provided for “free”, as opposed to “fee”. We're going to see a lot more companies figuring out creative ways of extracting more value out of data and then charging directly for that, rather than using that as an indirect way of generating traffic.

Gardner: We've seen examples of how this has been in place. Does it scale and does the governance or lack of governance that might be in the market now sustain us through the transition into Platform 3.0 and the Internet of Things.
Having standards is going to increasingly become important, unless we really address a lot of the data illiteracy that we have.

Gordon: That aspect is the lead-on part of “you get what you pay for”. If you're using a free source of data, you don't have any guarantee that it is from authoritative sources of data. Often, what we're getting now is something somebody put it in a blog post, and then that will get referenced elsewhere, but there was nothing to go back to. It's the shaky supply chain for data.

You need to think about the data supply and that is where the governance comes in. Having standards is going to increasingly become important, unless we really address a lot of the data illiteracy that we have. A lot of people do not understand how to analyze data.

One aspect of that is a lot of people expect that we have to do full population surveys, as opposed representative sampling to get much more accurate and much more cost-effective collection of data. That's just one example, and we do need a lot more in governance and standards.

Gardner: What would you like to see changed most in order for the benefits and rewards of the Internet of Things to develop and overcome the drawbacks, the risks, the downside? What, in your opinion, would you like to see happen to make this a positive, rapid outcome? Let's start with you Jean-Francois.

Barsoum: There are things that I have seen cities start to do now. There are couple of examples: Philadelphia is one and Barcelona does this too. Rather than do the typical request for proposal (RFP), where they say, "This is the kind of solution we're looking for, and here are our parameters. Can l you tell us how much it is going to cost to build," they come to you with the problem and they say, "Here is the problem I want to fix. Here are my priorities, and you're at liberty to decide how best to fix the problem, but tell us how much that would cost."

If you do that and you combine it with access to the public data that is available -- if public sector opens up its data -- you end up with a very powerful combination that liberates a lot of creativity. You can create a lot of new business models. We need to see much more of that. That's where I would start.

More education

Tabet: I agree with Jean-Francois on that. What I'd like to add is that I think we need to push the relation a little further. We need more education, to your point earlier, around the data and the capabilities.

We need these platforms that we can leverage a little bit further with the analytics, with machine learning, and with all of these capabilities that are out there. We have to also remember, when we talk about the Internet of Things, it is things talking to each other.

So it is not human-machine communication. Machine-to-machine automation will be further than that, and we need more innovation and more work in this area, particularly more activity from the governments. We've seen that, but it is a little bit frail from that point of view right now.

Gardner: Dave Lounsbury, thoughts about what need to happen in order to keep this on the tracks?
Thank you for mentioning the machine-to-machine part, because there are plenty of projections that show that it's going to be the dominant form of Internet communication, probably within the next four years.

Lounsbury: We've touched on lot of them already. Thank you for mentioning the machine-to-machine part, because there are plenty of projections that show that it's going to be the dominant form of Internet communication, probably within the next four years.

So we need to start thinking of that and moving beyond our traditional models of humans talking through interfaces to set of services. We need to identify the building blocks of capability that you need to manage, not only the information flow and the skilled person that is going to produce it, but also how you manage the machine-to-machine interactions.

Gordon: I'd like to see not so much focus on data management, but focus on what is the data managing and helping us to do. Focusing on the machine-to-machine and the devices is great, but it should be not on the devices or on the machines… it should be on what can they accomplish by communicating; what can you accomplish with the devices and then have a reverse engineer from that.

Gardner: Let's go to some questions from the audience. The first one asks about a high order of intelligence which we mentioned earlier. It could be artificial intelligence, perhaps, but they ask whether that's really the issue. Is the nature of the data substantially different, or we are just creating more of the same, so that it is a storage, plumbing, and processing problem? What, if anything, are we lacking in our current analytics capabilities that are holding us back from exploiting the Internet of Things?

Gordon: I've definitely seen that. That has a lot to do with not setting your decision objectives and your decision criteria ahead of time so that you end up collecting a whole bunch of data, and the important data gets lost in the mix. There is a term "data smog."

Most important

The solution is to figure out, before you go collecting data, what data is most important to you. If you can't collect certain kinds of data that are important to you directly, then think about how to indirectly collect that data and how to get proxies. But don't try to go and collect all the data for that. Narrow in on what is going to be most important and most representative of what you're trying to accomplish.

Gardner: Does anyone want to add to this idea of understanding what current analytics capabilities are lacking, if we have to adopt and absorb the Internet of Things?

Barsoum: There is one element around projection into the future. We've been very good at analyzing historical information to understand what's been happening in the past. We need to become better at projecting into the future, and obviously we've been doing that for some time already.

But so many variables are changing. Just to take the driverless car as an example. We've been collecting data from loop detectors, radar detectors, and even Bluetooth antennas to understand how traffic moves in the city. But we need to think harder about what that means and how we understand the city of tomorrow is going to work. That requires more thinking about the data, a little bit like what Penelope mentioned, how we interpret that, and how we push that out into the future.

Lounsbury: I have to agree with both. It's not about statistics. We can use historical data. It helps with lot of things, but one of the major issues we still deal with today is the question of semantics, the meaning of the data. This goes back to your point, Penelope, around the relevance and the context of that information – how you get what you need when you need it, so you can make the right decisions.
As soon as you talk about interoperability in the health sector, people start wondering where is their data going to go.

Gardner: Our last question from the audience goes back to Jean-Francois’s comments about the Canadian healthcare system. I imagine it applies to almost any healthcare system around the world. But it asks why interoperability is so difficult to achieve, when we have the power of the purse, that is the market. We also supposedly have the power of the legislation and regulation. You would think between one or the other or both that interoperability, because the stakes are so high, would happen. What's holding it up?

Barsoum: There are a couple of reasons. One, in the particular case of healthcare, is privacy, but that is one that you could see going elsewhere. As soon as you talk about interoperability in the health sector, people start wondering where is their data going to go and how accessible is it going to be and to whom.

You need to put a certain number of controls over top of that. What is happening in parallel is that you have people who own some data, who believe they have some power from owning that data, and that they will lose that power if they share it. That can come from doctors, hospitals, anywhere.

So there's a certain amount of change management you have to get beyond. Everybody has to focus on the welfare of the patient. They have to understand that there has to be a priority, but you also have to understand the welfare of the different stakeholders in the system and make sure that you do not forget about them, because if you forget about them they will find some way to slow you down.

Use of an ecosystem

Lounsbury: To me, that's a perfect example of what Marshall Van Alstyne talked about this morning. It's the change from focus on product to a focus on an ecosystem. Healthcare traditionally has been very focused on a doctor providing product to patient, or a caregiver providing a product to a patient. Now, we're actually starting to see that the only way we're able to do this is through use of an ecosystem.

That's a hard transition. It's a business-model transition. I will put in a plug here for The Open Group Healthcare vertical, which is looking at that from architecture perspective. I see that our Forum Director Jason Lee is over here. So if you want to explore that more, please see him.

Gardner: I'm afraid we will have to leave it there. We've been discussing the practical implications of the Internet of Things and how it is now set to add a new dimension to Open Platform 3.0 and Boundaryless Information Flow.
It's the change from focus on product to a focus on an ecosystem.

We've heard how new thinking about interoperability will be needed to extract the value and orchestrate out the chaos with such vast new scales of inputs and a whole new categories of information.

So with that, a big thank you to our guests: Said Tabet, Chief Technology Officer for Governance, Risk and Compliance Strategy at EMC; Penelope Gordon, Emerging Technology Strategist at 1Plug Corp.; Jean-Francois Barsoum, Senior Managing Consultant for Smarter Cities, Water and Transportation at IBM, and Dave Lounsbury, Chief Technology Officer at The Open Group.

This is Dana Gardner, Principal Analyst at Interarbor Solutions, your host and moderator throughout these discussions on Open Platform 3.0 and Boundaryless Information Flow at The Open Group Conference, recently held in Boston. Thanks again for listening, and come back next time.

Listen to the podcast. Find it on iTunes. Download the transcript. Sponsor: The Open Group

Transcript of a BriefingsDirect podcast exploring the challenges and ramifications of the Internet of Things, as machines and sensors collect vast amounts of data. Copyright The Open Group and Interarbor Solutions, LLC, 2005-2014. All rights reserved.

You may also be interested in:

Tuesday, June 24, 2014

The Open Group Amsterdam Conference Panel Delves into How to Best Gain Business Value From Open Platform 3.0

Transcript of a podcast from The Open Group Conference, exploring the future and direction of Open Platform 3.0.

Listen to the podcast. Find it on iTunesDownload the transcript. Sponsor: The Open Group.

Dana Gardner: Hello, and welcome to a special BriefingsDirect Podcast coming to you from a recent The Open Group Conference on May 13 in Amsterdam.

I’m Dana Gardner, Principal Analyst at Interarbor Solutions, and I’ll be your host for these discussions focused on  Enabling Boundaryless Information Flow.

Today, we present a series of presentations and a panel discussion on obtaining value from Open Platform 3.0, which is the shift to big datacloudInternet-of-Thingsmobile, and social in a combination that impacts business. Follow the ongoing discussions on twitter at #ogChat

First, I will hand it off to today's moderator, Stuart Boardman, a Senior Business Consultant at KPN and The Open Platform 3.0 Forum co-chairman. 

He'll be followed by Dr. Chris Harding, Director for Interoperability at The Open Group and The Open Platform 3.0 Forum Director, who will then hand it off to Lydia Duijvestijn, Executive Architect at IBM Global Business Services in the Netherlands. 

Later in the program, joining Stuart, Chris and Lydia, will be our additional panelists. They are: Andy Jones, Technical Director for EMEA at SOA Software; TJ Virdi, Computing Architect in the Systems Architecture Group at Boeing and also co-chair of the Open Platform 3.0 Forum; Louis Dietvorst, Enterprise Architect at Enexis in the Netherlands; Sjoerd Hulzinga, Charter Lead at KPN Consulting; and lastly, Frans van der Reep, Professor at the Inholland University of Applied Sciences. 

And now, here's our moderator, Stuart Boardman.

Stuart Boardman: Welcome to the first afternoon session about obtaining value from Open Platform 3.0, and how we're actually going to get value out of the things that we want to implement from big data, social, and the Internet-of-Things, etc., in collaboration with each other. 

Boardman
We're going to start off with Chris Harding, who is going to give us a brief explanation of what the platform is, what we mean by it, what we've produced so far, and where we're trying to go with it. 

He'll be followed by Lydia Duijvestijn, who will give us a presentation about the importance of non-functional requirements (NFRs). If we talk about getting business value, those are absolutely central. Then, we're going to go over to a panel discussion with additional guests. 

Without further ado, here's Chris Harding, who will give you an introduction to Open Platform 3.0. 

Purpose of architecture

Dr. Chris Harding: Hello, everybody. It's a great pleasure to be here in Amsterdam. I was out in the city by the canals this morning. The sunshine was out, and it was like moving through a set of picture postcards. 

Harding
It's a great city. As you walk through, you see the canals, the great buildings, the houses to the sides, and you see the cargo hoists up in the eaves of those buildings. That reminds you that the purpose of the arrangement was not to give pleasure to tourists, but because Amsterdam is a great trading city, that is a very efficient way of getting goods distributed throughout the city. 

That's perhaps a reminder to us that the primary purpose of architecture is not to look beautiful, but to deliver business value, though surprisingly, the two often seem to go together quite well. 

Probably when those canals were first thought of, it was not obvious that this was the right thing to do for Amsterdam. Certainly it would not be obvious that this was the right layout for that canal network, and that is the exciting stage that we're at with Open Platform 3.0 right now.

We have developed a statement, a number of use cases. We started off with the idea that we were going to define a platform to enable enterprises to get value from new technologies such as cloud computing, social computing, mobile computing, big data, the Internet-of-Things, and perhaps others.

We developed a set of business use cases to show how people are using and wanting to use those technologies. We developed an Open Group business scenario to capture the business requirements. That then leads to the next step. All these things sound wonderful, all these new technologies sound wonderful, but what is Open Platform 3.0? 

Jones
Though we don't have the complete description of it yet, it is beginning to take shape. That's what I am hoping to share with you in this presentation, our current thoughts on it. 

Looking historically, the first platform, you could say, was operating systems -- the Unix operating system. The reason why The Open Group, X/Open in those days, got involved was because we had companies complaining, "We are locked into a proprietary operating system or proprietary operating systems. We want applications portability." The value delivered through a common application environment, which was what The Open Group specified for Unix, was to prevent vendor lock-in. 

The second platform is the World Wide Web. That delivers a common services environment, for services either through accessing web pages for your browser or for web services where programs similarly can retrieve or input information from or to the web service. 

The benefit that that has delivered is universal deployment and access. Pretty much anyone or any company anywhere can create a services-based solution and deploy it on the web, and everyone anywhere can access that solution. That was the second platform. 

Common environment

The way Open Platform 3.0 is developing is as a common architecture environment, a common environment in which enterprises can do architecture, not as a replacement for TOGAF. TOGAF is about how you do architecture and will continue to be used with Open Platform 3.0. 

Open Platform 3.0 is more about what kind of architecture you will create, and by the definition of a common environment for doing this, the big business benefit that will be delivered will be integrated solutions. 

Yes, you can develop a solution, anyone can develop a solution, based on services accessible over the World Wide Web, but will those solutions work together out of the box? Not usually. Very rarely. 
There is an increasing need, which we have come upon in looking at The Open Platform 3.0 technologies. People want to use these technologies together. There are solutions developed for those technologies independently of each other that need to be integrated. That is why Open Platform 3.0 has to deliver a way of integrating solutions that have been developed independently. That's what I am going talk about. 

The Open Group has recently published its first thoughts on Open Platform 3.0, that's the White Paper. I will be saying what’s in that White Paper, what the platform will do -- and because this is just the first rough picture of what Open Platform 3.0 could be like -- how we're going to complete the definition. Then, I will wrap up with a few conclusions. 

So what is in the current White Paper? Well, what we see as being eventually in the Open Platform 3.0 standards are a number of things. You could say that a lot of these are common architecture artifacts that can be used in solution development, and that's why I'm talking about a common architecture environment.

Statement of need objectives and principles is not that of course; it's why we're doing it. 

Dietvorst
Definition of key terms: clearly you have to share an understanding of the key terms if you're going to develop common solutions or integrable solutions. 

Stakeholders and their concerns: an important feature of an architecture development. An understanding of the stakeholders and their concerns is something that we need in the standard. 

A capabilities map that shows what the products and services do that are in the platform. 

And basic models that show how those platform components work with each other and with other products and services. 

Explanation: this is an important point and one that we haven’t gotten to yet, but we need to explain how those models can be combined to realize solutions. 

Standards and guidelines

Finally, it's not enough to just have those models; there needs to be the standards and guidelines that govern how the products and services interoperate. These are not standards that The Open Group is likely to produce. They will almost certainly be produced by other bodies, but we need to identify the appropriate ones and, probably in some cases, coordinate with the appropriate bodies to see that they are developed.

van der Reep
What we have in the White Paper is an initial statement of needs, objectives, and principles; definitions of some key terms; our first-pass list of stakeholders and their concerns; and maybe half a dozen basic models. These are in an analysis of the use cases, the business use cases, for Open Platform 3.0 that were developed earlier. 

These are just starting points, and it's incomplete. Each of those sections is incomplete in itself, and of course we don't have the complete set of sections. It's all subject to change. 

This is one of the basic models that we identified in the snapshot. It's the Mobile Connected Device Model and it comes up quite often. And you can see, that stack on the left is a mobile device, it has a user, and it has a platform, which would probably be Android or iOS, quite likely. And it has infrastructure that supports the platform. It’s connected to the World Wide Web, because that’s part of the definition of mobile computing. 

On the right, you see, and this is a frequently encountered pattern, that you don't just use your mobile phone for running an app. Maybe you connect it to a printer. Maybe you connect it to your headphones. Maybe you connect it to somebody's payment terminal. You might connect it to various things. You might do it through a USB. You might do it through Bluetooth. You might do it by near field communication (NFC)
It's fundamental to mobile computing and also somewhat connected to the Internet of Things.

But you're connecting to some device, and that device is being operated possibly by yourself, if it was headphones; and possibly by another organization if, for example, it was a payment terminal and the user of the mobile device has a business relationship with the operator of the connected device.

That’s the basic model. It's one of the basic models that came up in the analysis of use cases, which is captured in the White Paper. As you can see, it's fundamental to mobile computing and also somewhat connected to the Internet-of-Things.

That's the kind of thing that's in the current White Paper, a specific example of all those models in the current White Paper. Let’s move on to what the platform is actually going to do? 

There are three slides in this section. This slide is probably familiar to people who have watched presentations on Open Platform 3.0 previously. It captures our understanding of the need to obtain information from these new technologies, the social media, the mobile devices, sensors, and so on, the need to process that information, maybe on the cloud, and to manage it, stewardship, query and search, all those things. 

Ultimately, and this is where you get the business value, it delivers it in a form where there is analysis and reasoning, which enables enterprises to take business decisions based on that information.

So that’s our original picture of what Open Platform 3.0 will do. 

IT as broker

This next picture captures a requirement that we picked up in the development of the business scenario. A gentleman from Shell gave the very excellent presentation this morning. One of the things you may have picked up him saying was that the IT department is becoming a broker.

Traditionally, you would have had the business use in the business departments and pretty much everything else on that slide in the IT department, but two things are changing. One, the business users are getting smarter, more able to use technology; and two, they want to use technology either themselves or to have business technologists closely working with them.

Systems provisioning and management is often going out to cloud service providers, and the programming, integration, and helpdesk is going to brokers, who may be independent cloud brokers. This is the IT department in a broker role, you might say. 

But the business still needs to retain responsibility for the overall architecture and for compliance. If you do something against your company’s principles, it's not a good defense to say, "Well, our broker did it that way." You are responsible. 
That's why we're looking for Open Platform 3.0 to define the common models that you need to access the technologies in question.

Similarly, if you break the law, your broker does not go to jail, you do. So those things will continue to be more associated with the business departments, even as the rest is devolved. And that’s a way of using IT that Open Platform 3.0 must and will accommodate. 

Finally, I mentioned the integration of independently developed solutions. This next slide captures how that can be achieved. Both of these, by the way, are from the analysis of business use cases. 

Also, you'll also notice they are done in ArchiMate, and I will give ArchiMate a little plug at this point, because we have found it actually very useful in doing this analysis. 

But the point is that if those solutions share a common model, then it's much easier to integrate them. That's why we're looking for Open Platform 3.0 to define the common models that you need to access the technologies in question.

It will also have common artifacts, such as architectural principles, stakeholders, definitions, descriptions, and so on. If the independently developed architectures use those, it will mean that they can be integrated more easily.

So how are we going to complete the definition of Open Platform 3.0? This slide comes from our business use cases’ White Paper and it shows the 22 use cases we published. We've added one or two to them since the publication in a whole range of areas: multimedia, social networks, building energy management, smart appliances, financial services, medical research, and so on. Those use cases touch on a wide variety of areas.

You can see that we've started an analysis of those use cases. This is an ArchiMate picture showing how our first business use case, The Mobile Smart Store, could be realized. 

Business layer

And as you look at that, you see common models. If you notice, that is pretty much the same as the TOGAF Technical Reference Model (TRM) from the year dot. We've added a business layer. I guess that shows that we have come architecturally a little way in that direction since the TRM was defined. 

But you also see that the same model actually appears in the same use case in a different place, and it appears all over the business use cases.

But you can also see there that the Mobile Connected Device Model has appeared in this use case and is appearing in other use cases. So as we analyze those use cases, we're finding common models that can be identified, as well as common principles, common stakeholders, and so on. 

So we have a development cycle, whereby the use cases provide an understanding. We'll be looking not only at the ones we have developed, but also at things like the healthcare presentation that we heard this morning. That is really a use case for Open Platform 3.0 just as much as any of the ones that we have looked at. We'll be doing an analysis of those use cases and the specification and we'll be iterating through that. 
This enables enterprises to derive business value from social computing, mobile computing, big data, the Internet-of-Things, and potentially new technologies. 

The White Paper represents the very first pass through that cycle. Further passes will result in further White Papers, a snapshot, and ultimately The Open Platform 3.0 standard, and no doubt, more than one version of that standard.

In conclusion, Open Platform 3.0 provides a common environment for architecture development. This enables enterprises to derive business value from social computing, mobile computing, big data, the Internet-of-Things, and potentially new technologies. 

Cognitive computing no doubt has been suggested as another technology that Open Platform 3.0 might, in due course, accommodate. What would that lead to? That would lead to additional use cases and further analysis, which would no doubt identify some basic models for common computing, which will be added to the platform. 

Open Platform 3.0 enables enterprise IT to be user-driven. This is really the revolution on that slide that showed the IT department becoming a broker, and devolvement of IT to cloud suppliers and so on. That's giving users the ability to drive IT directly themselves, and the platform will enable that. 

It will deliver the ability to integrate solutions that have been independently developed, with independently developed architectures, and to do that within a business ecosystem, because businesses typically exist within one or more business ecosystems. 

Those ecosystems are dynamic. Partners join, partners leave, and businesses cannot necessarily standardize the whole architecture across the ecosystem. It would be nice to do so, but by the time you finish the job, the business opportunity would be gone. 

So independently developed integration of independently developed architectures is crucial to the world of business ecosystems and delivering value within them. 

Iterative process

The platform will deliver that and is being developed through an iterative process of understanding the content, analyzing the use cases, and documenting the common features, as I have explained.

The development is being done by The Open Platform 3.0 Forum, and these are representatives of Open Group members. They are defining the platform. And the forum is not only defining the platform, but it's also working on standards and guides in the technology areas. 

For example, we have reformed a group to develop a White Paper on big data. If you want to learn about that, Ken Street, who is one of the co-chairs, is in this conference. And we also have cloud projects and other projects.

But not only are we doing the development within the Forum, we welcome input and comments from other individuals within and outside The Open Group and from other industry bodies. That’s part of the purpose of publishing the White Paper and giving this presentation to obtain that input and comment. 
The platform will deliver that and is being developed through an iterative process of understanding the content, analyzing the use cases, and documenting the common features

If you need further information, here's where you can download the White Paper from. You have to give your name and email address and have an Open Group ID and then it's free to download. 

If you are looking for deeper information on what the Forum is doing, the Forum Plato page, which is the next URL, is the place to find it. Nonmembers get some information there; Forum members can log in and get more information on our work in progress. 

If your organization is not a member of The Open Group, you can find out about Open Group membership from that URL. So thank you very much for your attention.

Boardman: Next is Lydia Duijvestijn, who is one of these people who, years ago when I first got involved in this business, we used to call Technical Architects, when the term meant something. The Technical Architect was the person who made sure that the system actually did what the business needed it to do, that it performed, that it was reliable, and that it was trustworthy. 

That's one of her preoccupations. Lydia is going to give us a short presentation about some ideas that she is developing and is going to contribute to The Open Platform 3.0. 

Quality of service

Lydia Duijvestijn: Like Stuart said, my profession is being an architect, apart from your conventional performance engineer. I lead a worldwide community within IBM for performance and competency. I've been working a couple of years with the Dutch Research Institute on projects around quality of service. That basically is my focus area within the business. I work for Global Services within IBM. 

Duijvestijin
What I want to achieve with this presentation is for you to get a better awareness of what functional requirements, functional characteristics, or quality of service characteristics are, and why they won't just appear out of the blue when the new world of Platform 3.0 comes along. They are getting more and more important. 

I will zoom in very briefly on three categories; performance and scalability, availability and business continuity, and security and privacy. I'm not going to talk in detail about these topics. I could do that for hours, but we don’t have the time. 

Then, I'll briefly start the discussion on how that reflects into Platform 3.0. The goal is that when we're here next year at the same time, maybe we would have formed a stream around it and we would have many more ideas, but now, it's just in the beginning.

This is a recap, basically, of a non-functional requirement. We have to start the presentation with that, because maybe not everybody knows this. They basically are qualities or constraints that must be satisfied by the IT system. But normally, it's not the highest priority. Normally, it's functionality first and then the rest. We'll find out about that later when the thing is going into production, and then it's too late. 

So what sorts of non-functionals do we have? We have run-time non-functionals, things that can be observed at run-time, such as performance, availability, or what have you. We also have non-run-time non-functionals, things that cannot apparently be tested, such as maintainability, but they are all very important for the system. 
Non-functionals are fairly often seen as a risk. If you did not pay attention to them, very nasty things could happen.

Then, we have constraints, limitations that you have to be aware of. It looks like in the new world, there are no limitations, cloud is endless, but in fact it's not true. 

Non-functionals are fairly often seen as a risk. If you did not pay attention to them, very nasty things could happen. You could lose business. You could lose image. And many other things could happen to you. It's not seen as something positive to work on it. It's seen as a risk if you don’t do it, but it's a significant risk. 

We've seen occasions where a system was developed that was really doing what it should do in terms of functionality. Then, it was rolled into production, all these different users came along, and the website completely collapsed. The company was in the newspapers, and it was a very bad place to be in. 

As an example, I took this picture in Badaling Station, near the Great Wall. I use this in my performance class. This depicts a mismatch between the workload pattern and the available capacity. 

What happens here is that you take the train in the morning and walk over to Great Wall. Then you've seen it, you're completely fed up with it, and you want to go back, but you have to wait until 3 o’clock for the first train. The Chinese people are very patient people. So they accept that. In the Netherlands people would start shouting and screaming, asking for better.

Basic mismatch

This is an example from real life, where you can have a very dissatisfied user because there was a mismatch between the workload, the arrival pattern, and available capacity. 

But it can get much worse, here we have listed a number of newspaper quotes as a result of security incidents. This is something that really bothers companies. This is also non-functional. It's really very important, especially when we go towards always on, always accessible, anytime, anywhere. This is really a big issue. 

There are many, many non-functional aspects, as you can see. This guy is not making sense out of it. He doesn’t know how to balance it, because it's not as if you can have them all. If you put too much focus on one, it could be bad for the other. So you really have to balance and prioritize. 

Not all non-functionals are equally important. We picked three of them for our conference in February: performance, availability and security. I now want to talk about performance. 
It's really very important, especially when we go towards always on, always accessible, anytime, anywhere. This is really a big issue. 

Everybody recognizes this picture. This was Usain Bolt winning his 100 meters in London. Why did I put this up? Because it very clearly shows what it's all about in performance. There are three attributes that are important.

You have the response time, basically you compare the 100 meters time from start to finish. 

You have the throughput, that is the number of items that can be processed with the time limit. If this is an eight-lane track, you can have only eight runners at the same time. And the capacity is basically the fact that this was an eight lane track, and they are all dependent on each other. It's very simple. But you have to be aware of all of them when you start designing your system. So this is performance. 

Now, let’s go to availability. That is really a very big point today, because with the coming of the Internet in the '90s, availability really became important. We saw that when companies started opening up their mainframes for the Internet, they weren't designed for being open all the time. This is about scheduled downtime. Companies such as eBay, Amazon, Google are setting the standard. 

We come to a company, and they ask us for our performance engineering. We ask them what their non-functional requirements are. They tell us that it has to be as fast as Google.

Well, you're not doing the same thing as Google; you are doing something completely different. Your infrastructure doesn’t look as commodity as Google's does. So how are you going to achieve that? But that is the perception. That is what they want. They see that coming their way.

Big challenge

They're using mobile devices, and they want it also in the company. That is the standard, and disaster recovery is slowly going away. RTO/RPO are going to 0. It's really a challenge. It's a big challenge.

The future is never-down technology independence, and it's very important to get customer satisfaction. This is a big thing.

Now, a little bit about security incidents. I'm not a security specialist. This was prepared by one of my colleagues. Her presentation shows that nothing is secure, nothing, and you have all these incidents. This comes from a report that tracks over several months what sort of incidents are happening. When you see this, you really get frightened. 

Is there a secure site? Maybe, they say, but in fact, no, nothing is secure. This is also very important, especially nowadays. We're sharing more and more personal information over the net. It's really important to think about this. 

What does this have to do with Platform 3.0? I think I answered it already, but let's make it a little bit more specific. Open Platform 3.0 has a number of constituents, and Chris has introduced that to you. 
In the Internet of Things,we have all these devices, sensors, creating huge amounts of data. They're collected by very many different devices all over the place. 

I want to highlight the following clouds, the ones with the big letters in it. There is Internet-of-Things, social, mobile, cloud, big data, but let’s talk about this and briefly try to figure out what it means in terms of non-functionals. 

In the Internet of Things,we have all these devices, sensors, creating huge amounts of data. They're collected by very many different devices all over the place. 

If this is about healthcare, you can understand that privacy must be ensured. Social security privacy is very important in that respect. And it doesn’t come for free. We have to design it into the systems. 

Now, big data. We have the four Vs there; Volume, Variety, Velocity, and Veracity. That already suggests a high focus on non-functionals, especially volume, performance, veracity, security, velocity, performance, and also availability, because you need this information instantaneously. When decisions have to be made based on it, it has to be there. 

So non-functionals are really important for big data. We wrote a white paper about this, and it's very highly rated. 

Cloud has a specific capacity of handling multi-tenant environments. So we have to make sure that the information of one tenant isn’t entered in another tenant’s environment. That's a very important security problem again. There are different workloads coming in parallel, because all these tenants have to have very specific types of workloads. So we have to handle it and balance it. That’s a performance problem. 

Non-functional aspects

Again, there are a lot of non-functional aspects. For mobile and social, the issue is that  you have to be always on, always there, accessible from anywhere. In social especially, you want to share your photos, you personal data, with your friends. So it's social security again. 

It's actually very important in Platform 3.0 and it doesn’t come for free. We have to design it into our model. 

That's basically my presentation. I hope that you enjoyed it and that it has made you aware of this important problem. I hope that, in the next year, we can start really thinking about how to incorporate this in Platform 3.0. 

Boardman: Let me introduce the panelists: Andy Jones of SOA Software, TJ Virdi from Boeing, Louis Dietvorst from Enexis, Sjoerd Hulzinga from KPN, and Frans van der Reep from Inholland University. 
The subject of interoperability, the semantic layer, is going to be a permanent and long running problem.

We want the panel to think about what they've just heard and what they would like Platform 3.0 to do next. What is actually going to be the most important, the most useful, for them, which is not necessarily the things we have thought of.

Andy Jones: The subject of interoperability, the semantic layer, is going to be a permanent and long running problem. We're seeing some industries. for example, clinical trials data, where they can see movement in that area. Some financial services businesses are trying to abstract their information models, but without semantic alignment, the vision of the platform is going to be difficult to achieve. 

Louis Dietvorst: For my vision on Platform 3.0 and what it should support, I am very much in favor of giving the consumer or the asking party the lead, empower them. If you develop this kind of platform thinking, you should do it with your stakeholders and not for your stakeholders. And I wonder how can we attach those kind of stakeholders that they become co-creators. I don’t know the answer. 

Male Speaker: Neither do I, but I feel that what The Open Group should be doing next on the platform is, just as my neighbor said, keep the business perspective, the user perspective, continuously in your focus, because basically that’s the only reason you're doing it. 

In the presentation just now from Lydia about NFRs, you need to keep in mind that one of the most difficult, but also most important, parts of the model ought to be the security and blind spots over it. I don’t disagree that they are NFRs. They are probably the most important requirements. It’s where you start. That would be my idea of what to do next. 

Not platform, but ecosystem

Male Speaker: Three remarks. First, I have the impression this is not a platform, but an ecosystem. So one should change the wording, number one.You should correct the wording. 

Second, I should stress the business case. Why should I buy this? What problem does it solve? I don’t know yet. 

The third point, as the Open Group, I would welcome a lobby to make IT vendors, in a formal sense, product reliable like other industries -- cars, for example. That will do a lot for the security problem the last lady talked about. IT centers are not reliable. They are not responsible. That should change in order to be a grownup industry. 

TJ Virdi: I agree about what’s been said, but I will categorize in three elements here what I am looking for from a Boeing perspective on what platform should be doing: how enterprises could create new business opportunities, how they can actually optimize their current business processes or business things, and how they can optimize the operational aspects. 

So if there is a way to expedite these by having some standardized way to do things, Open Platform 3.0 would be a great forum to do that. 
In the bottom layers, in the infrastructure, there is lot of reliability. Everything is very much known and has been developed for a long time.

Boardman: Okay, thanks.Louis made the point that we need to go to the stakeholders and find out what they want. Of course, we would love if everybody in the world were a member of The Open Group, but we realize that that isn’t going to be the case tomorrow, perhaps the day after, who knows. In the meantime, we're very interested in getting the perspectives of a wider audience. 

So if you have things you would like to contribute, things you would like to challenge us with, questions, more about understanding, but particularly if you have ideas to contribute, you should feel free to do that. Get in touch probably via Chris, but you could also get in touch with either TJ or me as co-chairs, and put in your ideas. Anybody who contributes anything will be recognized. That was a reasonable statement, wasn’t it Chris? You're official Open Group? 

Is there anybody down there who has a question for this panel, raise your hand? 

Duijvestijn: Your remark was that IT vendors are not reliable, but I think that you have to distinguish the layers of the stack. In the bottom layers, in the infrastructure, there is lot of reliability. Everything is very much known and has been developed for a long time. 

If you look at the Gartner reports about incidents in performance and availability, what you see is that most of these happen because of process problems and application problems. That is where the focus has to be. Regarding the availability of applications, nobody ever publishes their book rate.

Boardman: Would anybody like to react to that?

Male Speaker: I totally agree with what Lydia was just saying. As soon as you go up in the stack, that’s where the variation starts. That’s where we need to make sure that we provide some kind of capabilities to manage that easily, so the business can make those kind of expedited way to provide business solutions on that. That’s where we're actually targeting it. 

The lower in the stack we go, it's already commoditized. So we're just trying to see how far high we can go and standardize those things.

Two discussions

Male Speaker: I think there are two discussions together; one discussion is about the reliability on the total [IT process], where the fault is in a [specific IT stack]. I think that’s two different discussions.

I totally agree that IT, or at least IT suppliers, need to focus more on reliability when they get the service as a whole. The customers aren’t interested in where in the stack the problem is. It should be reliable as a whole, not on a platform or in the presentation layer. That’s a non-issue, non-operational, but a non-issue. The issue is it should be reliable, and I totally agree that IT has a long way to go in that department.  

Boardman: I'm going to move on to another question, because an interesting question came up on the Tweets. The question is: "Do you think that Open Platform 3.0 will change how enterprises will work, creating new line of business applications? What impact do you see?" An interesting question. Would anybody like to endeavor to answer that?

Male Speaker: That’s an excellent question actually. When creating new lines of business applications, what we're really looking for is semantic interoperability. How can you bridge the gap between social and business media kind of information, so you can utilize the concept of what’s happening in the social media? Can you migrate that into a business media kind of thing and make it a more agile knowledge or information transfer. 
We are seeing a trend towards line of business apps being composed from micro-apps. So there's less ownership of their own resources.

For example, in the morning we were talking about HL7 as being very heavyweight for healthcare systems. There may be need to be some kind of an easy way to transform and share information. Those kind of things. If we provide those kind of capabilities in the platform, that will make the new line-of-business applications easier to do, as well as it will have an impact in the current systems as well. 

Jones: We are seeing a trend towards line of business apps being composed from micro-apps. So there's less ownership of their own resources. And with new functionality being more focused on a particular application area, there's less utility bundling. 

It also leads on to the question of what happens to the existing line of business apps. How will they exist in an enterprise, which is trying to go for a Platform 3.0 kind of strategy? Lydia’s point about NFRs and the importance of the NFRs brings into light a question of applications that don’t meet NFRs which are appropriate to the new world, and how you retrofit and constrain their behavior, so that they do play well in that kind of architecture. This is an interesting problem for most enterprises. 

Boardman: There's another completely different granularity question here. Is there a concept of small virtualization, a virtual machine on a watch or phone? 

Male Speaker: On phones and all, we have to make a compartmentalized area, where it's kind of like a sandbox. So you can consider that as a virtualization of area, where you would be doing things and then tearing that apart. 

It's not similar to what virtualization is, but it's creating a sandbox in smart devices, where enterprises could utilize some of their functionality, not mingling up with what are called personal device data. Those things are actually part of the concept, and could be utilized in that way. 

Architectural framework

Question: My question about virtualization is linked to whether this is just an architectural framework. Because when I hear the word platform, it's something I try to build something on, and I don’t think this is something I build on. If you can, comment on the validity of the use of the word platform here. 

Male Speaker: I don’t care that much what it is called. If I can use it in whatever I am doing and it produces a positive outcome for me, I'm okay with it. I gave my presentation the Internet-of-Things, or the Internet of everything, or the everywhere or the Thing of Net, or the Internet of People. Whatever you want to call it, just name it, if you can identify its object that’s important to you. That’s okay with me. The same thing goes for Platform 3.0 or whatever.

I'm happy with whatever you want to call it. Those kinds of discussions don't really contribute to the value that you want to produce with this effort. So I am happy with anything. You don't agree?
What we're really trying to do is provide some kind of capabilities that would expedite enterprises to build their business solutions on that.

Male Speaker: A large part of architecture is about having clear understandings and what they mean.

Male Speaker: Let me augment what was just said, and I think Dr. Harding was also alluding to this. It is in the stage where we're defining what Platform 3.0 is. One thing for sure is that we're going to be targeting it as to how you can build that architectural environment. 

Whether it may have frameworks or anything is still to be determined. What we're really trying to do is provide some kind of capabilities that would expedite enterprises to build their business solutions on that. Whether it's a pure translation of a platform per se is still to be determined. 

Boardman: The Internet-of-Things is still a very fuzzy definition. Here we're also looking at fuzzy definitions, and it's something that we constantly get asked questions about. What do we mean by Platform 3.0? 

The reason this question is important, and I also think Sjoerd’s answer to it is important, is because there are two aspects of the problem. What things do we need to tie down and define because we are architects and what things can we simply live with. As long as I know that his fish is my bicycle, I'm okay. 

It's one of the things we're working on. One of the challenges we have in the Forum is what exactly are we going to try and tie down in the definition and what not? Sorry, I had to slip that one in. 

I wanted to ask about trust, how important you see the issue of trust. My attention was drawn to this because I just saw a post that the European Court of Justice has announced that Google has to make it possible for any person or organization who asks for it to have Google erase all information that Google has stored anywhere about them

I wonder whether these kinds of trust issues going to become critical for the success of this kind of  ecosystem, because whether we call it a platform or not, it is an ecosystem.

Trust is important

Male Speaker: I'll try to start an answer. Trust is a very important part ever since the Internet became the backbone of all of those processes and all of those systems in those data exchanges. The trouble is that it's very easy to compromise that trust, as we have seen with the word from the NSA as exposed by Snowden. So yes, trust ought to be a part of it, but trust is probably pretty fragile the way w're approaching it right now. 

Do I have a solution to that problem? No, I don't. Maybe it will come in this new ecosystem. I don't see it explicitly being addressed, but I am assuming that, between all those little clouds, there ought to be some kind of a trust relationship. That's my start of an answer.

Andy Jones: Trust is going to be one of those permanently difficult questions. In historical times, maybe the types of organizations that were highest in trust ratings would have been perhaps democratic governments and possibly banks, neither of which have been doing particularly well in the last five years in that area. 

It’s going to be an ethical question for organizations who are gathering and holding data on behalf of their consumers. We know that if you put a set of terms and conditions in front of your consumers, they will probably click on "agree" without reading it. So you have to decide what trust you're going to ask for and what trust you think you can deliver on. 
That data can then be summarized across groups of individuals to create an ensemble dataset. At what level of privacy are we then?

Data ownership and data usage is going to be quite complex. For example, in clinical trials data, you have a set of data, which can be identified against the named individual. That sounds quite fine, but you can then make that set of data so it’s anonymized and is known to relate to a single individual, but can no longer identify who. Is that as private? 

That data can then be summarized across groups of individuals to create an ensemble dataset. At what level of privacy are we then? It seems to quickly goes out of the scope of reason and understanding of the consumer themselves. So the responsibility for ethical behavior appears to lie with the experts, which is always quite a dangerous place.

Male Speaker: We probably all agree that trust management is a key aspect when we are converging different solutions from so many partners and suppliers. When we're talking about Internet of data, Internet-of-Things, social, and mobile, no one organization would be providing all the solutions from scratch. 

So we may be utilizing stuff from different organizations or different organizational boundaries. Extending the organizational boundaries requires a very strong trust relationship, and it is very significant when you are trying to do that.

Boardman: There was a question that went through a little while ago. I'm noticing some of these questions are more questions to The Open Group than to our panel, but one I felt I could maybe turn around. The question was: "What kind of guidelines is the Forum thinking of providing?"

I'd like to do is turn that around to the panel and ask: what do you think it would be useful for us to produce? What would you like a guideline on, because there would be lots of things where you would think you don’t need that, you'll figure it out for yourself. But what would actually be useful to you if we were to produce some guidelines or something that could be accepted as a standard? 

Does it work?

Male Speaker: Just go to a number of companies out there and test whether it works. 

Male Speaker: In terms of guidelines, you mentioned it very well about semantic interoperability. How do you exchange information between different participants in an ecosystem or things built on a platform. 

The other thing is how you can standardize things that are yet to be standardized. There's unstructured data. There are things that need to be interrogated through that unstructured data. What are the guiding principles and guidelines that would do those things? So maybe in those areas, Platform 3.0 with the participations from these Forum members, can advance and work on it. 

Andy Jones: I think contract, composition, and accumulation. If an application is delivering service to its end users by combining dozens of complementary services, each of which has a separate contract, what contract can it then offer to its end user?

Boardman: Does the platform plan to define guidelines and directions to define application programming interfaces (APIs) and data models or specific domains? Also, how are you integrating with major industry reference models? 

Just for the information, some of this is work of other parts of The Open Group's work around industry domain reference models and that kind of thing. But in general, one of the things we've said from the Platform, from the Forum, is that as much as possible, we want to collate what is out there in terms of standards, both in APIs, data models, open data, etc.
No single organization would be able to actually tap into all the advancement that’s happening in technologies, processes, and other areas where business could utilize those things so quickly.

We're desperate not to go and reproduce anybody else’s work. So we are looking to see what’s out there, so the guideline would, as far as possible, help to understand what was available in which domain, whether that was a functional domain, technical domain, or whatever. I just thought I would answer those because we can’t really ask the panel that.

We said that the session would be about dealing with realizing business value, and we've talked around issues related to that, depending on your own personal take. But I'd like to ask the members of the panel, and I'd like all of you to try and come up with an answer to it: What do you see are the things that are critical to being able to deliver business value in this kind of ecosystem?

I keep saying ecosystem, not to be nice to Frans, I am never nice to Frans, but because I think that that captures what we are talking about better. So do you want to start TJ? What are you looking for in terms of value? 

Virdi: No single organization would be able to actually tap into all the advancement that’s happening in technologies, processes, and other areas where business could utilize those things so quickly. The expectations from business values or businesses to provide new solutions in real-time, information exchange, and all those things are the norm now. 

We can provide some of those as a baseline to provide as maybe foundational aspects to business to realize those new things what we are looking as in social media or some other places, where things are getting exchanged so quickly, and the kind of payload they have is a very small payload in terms of information exchange.

So keeping the integrity of information, as well as sharing the information with the right people at the right time and in the right venue, is really the key when we can provide those kind of enabling capabilities.

Ease of change

Andy Jones: In Lydia’s presentation, at the end, she added the ease of use requirement as the 401st. I think the 402nd is ease of change and the speed of change. Business value pretty much relies on dynamism, and it will become even more so. Platforms have to be architected in a way that they are sufficiently understood that they can change quickly, but predictably, maintaining the NFRs. 

Louis Dietvorst: One of the reasons why I would want to adopt this new ecosystem is that it gives me enough feeling that it is a reliable product. What we know from the energy system innovations we've done the last three or four years is that the way you enable and empower communities is to build up the trust themselves, locally, like you and your neighbor, or people who are close in proximity. Then, it’s very easy to build trust. 

Some call it social evidence. I know you, you know me, so I trust you. You are my neighbor and together we build a community. But the wider this distance is, the less easy it is to trust each other. That’s something you need to build in into the whole concept. How do you get the trust if it is something that's a global concept. It seems hardly possible.

Frans van der Reep: This ecosystem, or whatever you're going to call it, needs to bring the change, the rate of change. "Change is life" is a well-known saying, but lightning-fast change is the fact of life right now, with things like social and mobile specifically. 

One Twitter storm and the world has a very different view of your company, of your business. Literally, it can happen in minutes. This development ought to address that, and also provide the relevant hooks, if you will, for businesses to deal with that. So the rate of change is what I would like to see addressed in Platform 3.0, the ecosystem. 
In order to create meaningful customer interaction, what we used to call center or whatever, that is where the cognition comes in.

Male Speaker: It should be cheap and reliable, it should allow for change, for example Cognition-as-a-Service, and it should hide complexity for those "stupid businesspeople" and make it simple. 

Boardman: I want to pick up on something that Frans just said because it connects to a question I was going to ask anyway. People sometimes ask us why the particular five technologies that we have named in the Forum: cloud, big data, big-data analysis, social, mobile, and the Internet-of-Things? It's a good question, because fundamental to our ideas in the Forum that it’s not just about those five things. Other things can come along and be adopted. 

One of the things that we had played with at the beginning and decided not to include, just on the basis of a feeling about lack of maturity, was cognitive computing. Then, here comes Frans and just mentions cognitive things. 

I want to ask the panel: "Do you have a view on cognitive computing? Where is it? When we can expect it to be something we could incorporate? Is it something that should be built into the platform, or is it maybe just tangential to the platform?" Any thoughts? 

Male Speaker: I did a speech on this last week. In order to create meaningful customer interaction, what we used to call center or whatever, that is where the cognition comes in. That's a very big market and there's no reason not to include it in the lower levels of the platform and to make it into cloud. 

We have lots of examples already in the Netherlands that ICT devices recognize emotions and from recognizing speech. Recognizing emotion, you can optimize the matching of the company with the customer, and you can hide complexity. I think there’s a big market for that. 

What the business wants

Virdi: We need to look at it in the context of what business wants to do with that. It could be enabling things that could be what I consider as proprietary things, which may not be part of the platform for others to utilize. So we have to balance out what would be the enabling things we can provide as a base of foundation for everyone to utilize. Or companies can build on top of it what values it would provide. We probably have to do a little bit further assessment on that.

Male Speaker: I'd like to follow up on this notion of cognitive computing, the notion that maybe objects are self-aware, as opposed to being dumb -- self-aware being an object, a sensor that’s aware of its neighbor. When a neighbor goes away, it can find other neighbors. Quite simple as opposed to a bar code. 

We see that all the time. We have kids that are civil engineers and they pour it in concrete all the time. In terms of cost, in terms of being able to have the discussion, it's something that’s in front of us all the time. So at this time, should we probably think about at least the binary aspect of having self-aware sensors as opposed to dumb sensors?

Male Speaker: From aviation perspective, there are some areas where dumb devices would be there, as well as active devices. There are some passive sensor devices where you can just interrogate them when you request and there are some devices that are active, constantly sending sensor messages. Both are there in terms of utilization for business to create new business solutions. 
I'm certainly all in favor of devices in the field being able to tell you what they're doing and how they think they're feeling.

Both of them are going to be there, and it depends upon what business needs are to support those things. Probably we could provide some ways to standardize some of those and some other specifications. For example, an ATA, for aviation. They're doing that already. Also, in healthcare, there's HL7, looking for doing some smart sensor devices to exchange information as well. So some work is already happening in the industry. 

There are so many business solutions that have already been built on those. Maybe they're a little bit more proprietary. So a platform could provide some ways to provide a standard base to exchange that information. It may be some things relate to guidelines and how you can exchange information in those active and passive sensor devices.

Andy Jones: I'm certainly all in favor of devices in the field being able to tell you what they're doing and how they think they're feeling. I have an interest in complex consumer devices in retail and other field locations, especially self-service kiosks, and in that field quite a lot of effort has been spent trying to infer the states of devices by their behavior, rather than just having them tell you what's going on, which should be so much easier. 

Male Speaker: Of course, it depends on where the boundary is between aware and not aware. If there is thermometer in the field and it sends data that it's 15 degrees centigrade, for example, do I really want to know whether it thinks it's chilly or not? I'm not really sure about it. 

I'd have to think about it a long time to get a clear answer on whether ther's a benefit in self-aware devices in those kinds of applications. I can understand that there will be an advantage in self-aware sensor devices, but I struggle a little to see any pattern or similarities in those circumstances. 

I could come up with use cases, but I don’t think it's very easy to come up with a certain set of rules that leads to the determination whether or not a self-aware device is applicable in that particular situation. It's a good question. I think it deserves some more thought, but I can't come up with a better answer than that right now.

Case studies

Mark Skilton: I just wanted to add to the embedded question, because I thought it was a very good one. Three case studies happened to me recently. I was doing some work with Rolls Royce and the MH370, the flight that went down. One of the key things about the flight was that the engines had telemetry built in. TJ, you're more qualified to talk about this than I am, but essentially there was information that was embedded in the telemetry of the technology of the plane. 

As we know from the mass media that reported on that, that they were able to analyze from some of the data potentially what was going on in the flight. Clearly, with the band connection, it was the satellite data that was used to project it was going south, rather than north. 

So one of the lessons there was that smart information built into the object was of value. Clearly, there was a lesson learned there. 

With Coca Cola, for example, what's very interesting in retail is that a lot of the shops now have embedded sensors in the cooler systems or into products that are in the warehouse or on stock. Now, you're getting that kind of intelligence over RFID coming back into the supply chain to do backfilling, reordering, and stuff like that. So all of this I see is smart. 
Embedded technology in the dashboard is going to be something that is going to be coming in the next three to five years.

Another one is image recognition when you go into a car park court. You have your face being scanned in, whether you want it or not. Potentially, they can do advertising in context. These are all smart feedback loops that are going on in these ecosystems and are happening right now. 

There are real equations of value in doing that. I was just looking at the Open Automotive Alliance. We've done some work with them around connected car forecast. Embedded technology in the dashboard is going to be something that is going to be coming in the next three to five years with BMW, Jaguar Land Rover, and Volvo. All the major car players are doing this right now. 

So Open Platform 3.0 for me is riding that wave of understanding where the  intelligence and the feedback mechanisms work within each of the supply chains, within each of the contexts, either in the plane, in the shop, or whatever, starting to get intelligence built in. 

We talk about big data and small data at the university that I work at. At the moment, we're moving from a big-data era, which is analytics, static, and analyzing the process in situ. Most likely it's Amazon sort of purchasing recommendations or advertisement that you see on your browser today. 

We 're moving to a small-data era, which is where you have very much data in context of what's going on in the events at that time. I would expect this with embedded technologies. The feedback loops are going to happen within each of the traditional supply chains and will start to build that strength.

The issue for The Open Group is to capture the sort of standards of interoperability and connectivity much like what Boeing is already leading with, with the automotive sector , and with the airline sector. It's riding that wave, because the value of bringing that feedback into context, the small-data context is where the future lies. 

Infrastructure needed

Male Speaker: I totally agree. Not only are the devices or individual components getting smarter, but that requires infrastructures to be there to utilize that sensing information in a proper way. From the Platform 3.0 guidelines or specifications perspective, determining how you can utilize some devices, which are already smart, and others, which are still considered to be legacy, and how you can bridge those gap would be a good thing to do.

Boardman: Would anyone like to add anything, closing remarks?

Andy Jones: Everybody’s perspective and everybody’s context is going to be slightly different. We talked about whether it's a platform ora framework. In the end there will be a built universal 3.0 Platform, but everybody will still have a different view and a different perspective of what it does and what it means to them. 
My suggestion would be that, if you're going to continue with this ecosystem, try to built it up locally, in a locally controlled environment.

Male Speaker: My suggestion would be that, if you're going to continue with this ecosystem, try to built it up locally, in a locally controlled environment, where you can experiment and see what happens. Do it at many places at the same time in the world, and let the factors be proof of the pudding. 

Male Speaker: Whatever you are going to call it, keep to 3.0, that sounds snappy, but just get the beneficiaries in, get the businesses in, and get the users in.

Male Speaker: The more open, the more a commodity it will be. That means that no company can get profit from it. In the end, human interaction and stewardship will enter the market. If you come to London city airport and you find your way in the Tube, there is a human being there who helps you into the system. That becomes very important as well. I think you need to do both, stewardship and these kinds of ecosystems that spread complexity. 

Boardman: That's it for this session. I'd like to ask your applause for our panel and also our speakers.

Gardner: You've been listening to a special BriefingsDirect Podcast coming to you from The Open Group Conference on May 13 in Amsterdam. 

We've heard a series of presentations and a panel discussion, as well as a question-and-answer session, all on obtaining value from Platform 3.0. 

So a big thank you to our contributors here today: Stuart Boardman, a Senior Business Consultant at KPN and Open Platform 3.0 Forum co-chairman; Dr. Chris Harding, Director for Interoperability at The Open Group and Open Platform 3.0 Forum Director; Lydia Duijvestijn, Executive Architect at IBM Global Business Services; Andy Jones, Technical Director for EMEA at SOA Software; TJ Virdi, Computing Architect at Boeing and also a co-chair of The Open Platform 3.0 Forum; Louis Dietvorst, Enterprise Architect at Enexis; Sjoerd Hulzinga, Charter Lead at KPN Consulting; and lastly, Frans van der Reep, Professor at Inholland University of Applied Sciences. 

And of course a big thank you to our audience for joining this special podcast presentation. This is Dana Gardner, Principal Analyst at Interarbor Solutions, your BriefingsDirect host for this podcast. Thanks again for listening and come back next time.

Listen to the podcast. Find it on iTunesDownload the transcript. Sponsor: The Open Group.

Transcript of a podcast from The Open Group Conference, exploring the future and direction of Open Platform 3.0. Copyright The Open Group and Interarbor Solutions, LLC, 2005-2014. All rights reserved.

You may also be interested in: