Gregory Maciag

Short IoT Window? Not Without Standards

Are businesses taking IoT seriously? You bet. How about this: “A 90-year-old industrial device manufacturer that mFrontiers works with estimates that it has four years to make its analog devices more digitally savvy. They need to be internet-connected so that they can generate data on performance and monitor it remotely. After that, [mFrontiers'] Ottoman says, the company thinks internet-connected devices will be the standard, and vendors that don’t offer IoT functionality won’t win new business.”

This company is looking at a four year window. It's a make-or-break time period. And it's really short.  Forbes

Data Modeling is Data Governance

This is the title of an article and webinar by Robert S. Seiner. It's designed to ruffle feathers, but Seiner backs up the assertion.  For me, the clincher is the negative argument Seiner mentions – this is: “Ask any organization that does not model their data if their data is being governed. The sure answer will be 'no'.”

Data modeling is about defining data, and data governance is about managing data. If you haven't defined it, you can't manage it.  Note that data modeling doesn't automatically mean defining data from scratch. Organizations which use industry models as the basis for their own data models are exercising the same kind of discipline – while saving time and money.

So, if you want to be picky, I guess you might want to say: Data modeling is essential for data governance. But it doesn't make for such a great slogan.   TDAN

Data Management in China

The Chinese have noticed the DATA Act. China plans to create a unified open data platform for government by 2018. They'll have joined-up information across the state and one-stop citizen services.    Any doubt that data management is a vital aspect of national infrastructure ought to be dispelled by this news.   China

Backtracking to Standards

There's a familiar pattern to a recent article about interoperability in military robotics and autonomous systems: “Despite substantial increases in capability and applications, U.S. and multinational robotics and autonomous systems have limited information interoperability, convoluting an already complex data-sharing environment. The U.S. Defense Department finds itself in a predicament created by rapid and independent fielding of systems over the past 10 to 15 years along with the use of proprietary software and payload and bandwidth restrictions.”

We see this in business too. Organizations can't wait for data standards – they need to execute today.  Effectiveness today generates ineffectiveness for tomorrow.

There's no way out of this dilemma. If data standards don't exist, you have to go with what you have. It's good to be honest about the remedial work you'll need to do farther down the pike. It's also sensible to set a standards development activity in progress alongside the development of point solutions.

The development of industry data standards means many sectors, like insurance, don't need to take this kind of risk. I hope the close involvement of insurance players with the development of IoT will mean the commercial sector does not repeat the experience of the military with robotics and autonomous systems. Our industry needs to take a strong lead.

The article which inspired this post is fairly dense and technical. But there's a catchy nugget of business insight there too. The writers observe: “Data is to information as sand is to silicon.”    AFCEA

Information Blocking

The terms of debate in the realm of data management are undergoing radical change. With data at the heart of so many activities – not just in business, but in government as well – we're seeing a shift away from arguing about the value of data. This argument has been won. The friction now is about impeding the flow of data. Flow is now a given: obstruction is a problem.

This is clear in a recent piece about “information blocking” in healthcare. Are EHR (electronic health record) vendors blocking the exchange of data across the value chain? Or is this a simplistic attempt to assign blame for problems in an inherently complex business area?

Whatever the rights and wrongs in the case, we need to notice that data flow is being seen here as a kind of right in itself – or as a natural feature of collaborating organizations. It's a great step forward to see things in this way. Every multi-party endeavor must have unimpeded, secure, and meaningful data flow. Otherwise the value of the whole will be even less than the sum of the parts.    Health IT

Loose Standards

Is a “loose data standard” a contradiction in terms?

The idea of “loose or even ad hoc” standards for metadata pops up in an interesting post by Charlie Greenberg. One of his fictional examples has the term “Retired Employees” defined in different places as “age >65 years” and “age 65-79 years”. This would lead to different counts in different parts of the organization – and perhaps even truncation of benefits to some people of 80 and over.   Is that a loose definition? I think it's a case of a faulty standard. Because it's not standard.

But I think the idea of loose standards does have some merit. This is where data items are optional. The optionality of data items, and the logic which can group certain optional items together, is an important aspect of the business. For example, if you're insuring home contents, it's optional for people to have a bicycle. But if they have a bicycle, you want to know whether it's the kind of bicycle people like to steal.

Creating standards around optional items needs the kind of looseness engineers call tolerance – or, less formally, wiggle room. This is why we talk about standards being robust and resilient, rather than rigid. They need to be created with flexibility in mind.   INFOSEC

Open Data Gets Joined Up

The launch of the International Open Data Charter in New York is a welcome step forward for the data management community as well as everyone concerned with the open data movement. This is the first time we've seen a proposed data standards approach for everything and everywhere – that is, an attempt to enable data access across existing boundaries.

Also, as Bill Anderson says in his report on the Charter: “the inclusion of the principle of 'Comparability and Interoperability' [is] the first time that a charter like this has moved beyond transparency to properly consider the whole point of open data: usage. Most datasets are pretty meaningless on their own. It is only when they are combined and contextualized do they generally make sense. Joined-up data creates the information we need.”

There's long been a problem in the field of open data around the idea that just publishing something is good enough. It's finally getting through to more people that open data needs to be usable – and that means it must conform to some declared set of data definitions.    Open Data

Making Data Interesting

Socrata's Q&A with Tyrone Grandison, Deputy Chief Data Officer for the U.S. Department of Commerce, includes some wonderful advice on “making open data interesting”: “Whenever possible, tie it to a standard. Always ensure that it meets one or more use cases. Provide an example of a story that can be told with the open data you are creating.”

I believe this advice can be extended to the corporate setting as well. We need to keep humanizing the value of data and relating data to processes and outcomes people care about.

Note how and continues into stories. It's as if you need data standards as Grandison's advice begins with standardsthe foundation for the picture you're going to paint about the usefulness of the data. Even if you're only going to produce one case, one story, you still want to be sure – whenever possible – that the data you're talking about has multiple uses in different scenarios, and data standards are about making sure of exactly that.

From the perspective of commerce, I don't think the open data movement has too much to worry about when it comes to generating interest. There's a natural enthusiasm for open data which is expressed by the great levels of volunteering we see in open data, and the growth of related start-up activity. We could do with more of this spirit inside our corporations. And I believe it's coming – the zest for data is leaking into organizations and the excitement about what can be achieved with data is continuing to grow.   Socrata

Blockchain

The Securities and Exchange Commission has approved a plan from online retailer Overstock.com to issue company stock via the Internet, signaling a significant shift in the way financial securities will be distributed and traded in the years to come.    SEC

 

Securing/Exploiting Data with Standards

Data isn't a systems concern any more. Data is everybody's business. Going forward, I believe organizations' investments in data will be mostly in two areas: securing data, and exploiting data. Data standards are at the heart of both areas.

To insure all your data is secure, you need to know you're addressing all the relevant data types before you start applying a security regime. You can have the safest, most bullet-proof security systems on the planet – but if some of your crucial business information has been left outside of the umbrella, you're in trouble.

You think you're covered but you're not. Many organizations are uncertain about how much important corporate data is exposed to threats by virtue of its not being known about by data management. Using data standards – and continuously assessing their use – enables you to bring all your data under one management canopy. You can then be certain that your security regime is covering all your data.

In order to exploit data successfully, you need to be sure you're comparing like with like. Data standards are there to insure your information is meaningful, so you can take decisions based on analysis with confidence.

The original motivation for standards – the exchange of information between systems and business partners – hasn't gone away. It's still a big driver for adoption and an obvious entry point into measuring the business benefits of standards.

This motivation is absent in the cases of data security and data exploitation. Organizations may only appreciate the benefit of standards for security when they suffer a breach. They may only recognize how standards can help them make informed decisions after they make a damaging decision based on faulty data.

It's on leaders to look ahead to these outcomes and make sure they don't happen. Twenty-twenty hindsight isn't much use if you're using it from the vantage point of the scrap heap. To keep your data secure, and to leverage its incredible value, you need standards.

Learning from Food Safety Standards

The creation of business data standards is spreading out from commercial and government organizations to the big wide world. The motivations for new standards, and the methods being used to formulate them, are instructive. Looking at how these new standards movements are faring may well help business organizations to improve their own standards processes – saving time and money, and opening up new potential benefits.

A paradigm case for the new, in-the-wild standards world is that of food inspection. Food inspection could sound like something dull and technical – a box-checking exercise designed to add costs to restaurateurs. But badly sourced or prepared food has major health impacts. An e-coli outbreak in 1993 claimed more than 700 victims in four states. And this wasn't just a few upset stomachs: over 170 victims were left with permanent kidney and brain damage, and four children died. It was traced to under-cooked burger patties.

Data standards for this domain are a two-level job. The standards must define the key data elements from the ground up as well as ensure compliance with the definitions. In business, it's been usual for the first level of this task to be reasonably straightforward. Say you want to define a “product”. You can usually come to an agreement about what a product is, its attributes and relationships, even if there's a deal of discussion and compromise along the way. The data associated with food safety is much more diverse and there's less of a common mental model waiting to be dug out of the participants. And, which is very interesting, a large part of food safety monitoring comes from the experiences and impressions of customers.

Now, the challenges facing the creators of food safety standards look very much like those facing businesses in an age of big data and social media. Business isn't fully in control of the data model any more. Consumers, partners, and regulators contribute as much – if not more – to the model than the business entities who think they “own” the domain.

What can standards advocates and adopters in business learn from areas like food safety? First, remind people of the stakes. Bad food kills people – get the data right and you save lives. Characterize the risks of bad data in your own domain, whether those are lost sales, reduced market share, slow response, rework costs, reputational damage, blocked access to markets etc.

Second, expect to discuss data items that are completely new to the domain. Many of these will come from the Internet of Things, initially from wearable tech. What do we want the sensors in our environment to measure? What scales should these measurements use? What are the threshold points we're interested in?

And third – solicit and embrace data from other parties. A diner's assessment of how clean a room is matters to other diners, and therefore ultimately to the restaurant's owners. Your customers' feedback is not just useful for informing how you develop the business. It's also a primary source for other customers' behavior. Everything you do is now public.    Government Technology

Data Standards and Cloud Service Providers

Much of the innovation in farming today is around what they call precision agriculture. Precision ag means using data across farming processes to manage the business. It's like bringing industrial control to agriculture. You could also say it's about bringing real-time, data-based management to farming.

There are many precision ag products and services out there. But: “Farmers and others agree on the need for some sort of industry body that will define and govern the standards our industry should follow when it comes to data platforms.” There are two groups working on standards, one which seems to come at the problem from the systems point of view and one that runs under an open data banner. Of course these two viewpoints are not incompatible – they could just represent different starting points, one inspired by current offerings from vendors and the other by users.

The pattern of an industry becoming more centered on data and then seeking to build out standards is familiar from other domains. I guess the difference industries are dealing with today, in comparison with even five years ago, is the growth of cloud services. It's easier to sign up with a cloud service than to trial an out-of-the-box solution. The economics look a whole lot better and there's less apparent risk of lock-in. However, until a community agrees its core business data standards, users of these services face potential difficulties with shifting their data from one service provider to another.

I'd like cloud service providers to take a more proactive approach to industry standards. I know this hardly looks like it would be in their interests. They want to offer agnostic services and they're not in the business of helping customers move to other suppliers. But as the cloud services business matures, customers will be looking for value-add. It doesn't cost much for a cloud service provider to say “we recommend and support the industry data standards from XYZ”. Doing so would show some added customer orientation on their part. Most importantly, promotion of industry standards grows the entire business for everyone involved. There's market share – and there's helping to grow the market.    Ag Industry

Industry Innovation from ACORD

Bill Hartnett, ACORD Head of Innovation was interviewed in a podcast from Insurance Journal:

IJ COMMENT:  "It is easy to talk about disruption and the need for innovation. It is quite another thing to actually implement new and innovative processes, products and servides. ACORD recognizes this and under Bill Hartnett's leadership (as Head of Innovation) is challenging carriers, agents and brokers, associations, and solution providers to become innovative brands and drive innovation in our industry.

Bill recognizes that the insurance industry has always talked about innovation but historically implementation has been lacking. According to Bill, what is different today is that instead of the industry setting its own agenda it is being set from outside the industry. That is, there is external pressure from consumers being placed on our industry to innovate and adapt. Are you an innovative brand? Bill Hartnett and ACORD would love to hear from you."

Here's the link: Insurance Journal

Data Agnostic

This is a good quote from Chuck Helveston at Archer, over in the investment management business: “Until the SMA [separately managed account] industry universally adopts standards for a common global language, any system supporting SMAs will need to be data agnostic.”

This is sound advice for buyers of systems. But it's also kind of sad for vendors. To be data agnostic, you have to “believe in” every proprietary data standard in your industry and be able to translate between them. This adds development and maintenance overhead to systems. So it's bad for buyers of systems as well, who end up paying higher prices. By delaying agreement on industry standards, everybody suffers.  ARCHER

Data Skills Everywhere

I keep reading that organizations need to hire more people with skills in big data analytics and informatics. It's probably true. But who are these people, and where are they going to come from?

For the insurance industry, I see a close overlap between actuarial, underwriting and analytics skills. They're not the same functions, but they speak to similar interests and aptitudes.

But I believe the spread of technology must inevitably enable more people to realize they have an affinity for data analysis – people who otherwise would not have recognized their calling. Our stereotyped ideas about the kinds of people who are interested in this stuff and are good at it are probably outdated. Especially with the great advances in visual modeling. You no longer need to have some mysterious ability to “see” data patterns or relationships in your head. 

Also, it's vital to remember data is central to the business. We should not be corralling “data scientists” off in a sealed room and thinking of what they do as magic. Data analysis needs to be located in, and performed by, the business. Let's look for data analysis potential in everyone we recruit. And let's offer training and encouragement to everyone who shows an interest in working with data.   Health IT

Some Proprietary Standards are Harmless like DUNS

Here's a small mystery. At the end of August 2015, federal government officials finalized 57 data standards enabling the DATA Act. GitHub – the open source web platform which hosted much of the discussion about the formation of the standards – said: “Notably, Treasury and OMB chose to adopt the proprietary DUNS Number as the favored, government-wide identifier for recipients of federal funds – even though every GitHub comment said they shouldn't.”

What could the thinking be here? Maybe it's that the DUNS (Data Universal Numbering System) number is so well entrenched in existing systems that people see no reason to change to something else. If you're looking to standardize, wouldn't you want to preserve any pre-existing standard used across the domain and regard it as a time saving?

I guess the issue here is whether you see the proprietary nature of the DUNS number as a problem or not. For straight serial numbers, I'm not sure I care whether a private or public organization allocates the numbers. It's not as if they can lock anyone out.

I can also see how the adoption of any proprietary element could look like a precedent for favoring proprietary standards in general. But I don't think standards leaders think like this. Using the DUNS number doesn't mean government will lean toward other proprietary standards that potentially distort competition. It's not in anyone's interest.   FIERCE