Gregory Maciag

It's All about the DATA

Great quote from Pam Baker: “The Internet of Things isn’t about the things, it’s about the data.”

Baker shrewdly reviews the ongoing explosion in healthcare data, and warns that the volume and variety of data is only going to grow. She urges greater attention to proactive data management to avoid fragmentation and silos. She takes a broad view but the common thread is data standards: “It’s time for the entire medical and research community worldwide to come together, instead of working alone or in small groups, to address standardizing data from devices and fabrics and integrating data bases.”   NUVIUN

Evolving an Imortant Standard

HTTP/2 has been finalized and is set to be the new standard for creating web pages. This is the first update to the standard since HTTP 1.1 in 1999.

The web has grown not just in size but importance over the intervening time period. Developers have had to work around a number of limitations, especially in speed of page delivery. Why did it take so long to bring out a new version?

My take is that, whatever the performance issues around HTTP in our massively online world, it worked to an acceptable level. Changing such a crucial and entrenched standard is not something you do lightly. Also, I notice that HTTP/2 doesn't replace existing APIs, but adds features to them. The new version of the standard retains the core functionality so existing investments are not wiped out.

It's a great sign of HTTP's strength – and its core simplicity – that the world has been able to rely on HTTP 1.1 for so long. The grace with which the standard has been advanced to the new version is a tribute to a reasoned, collaborative standards setting process.   TNW

Standards are Tools

I lost the link but... Debra Walton of Thomson Reuters asks a great question: “Given that data is such a strategic asset, one of the biggest issues is understanding your infrastructure, your data standards and data governance. […] How do you have sufficient standards but create an environment that’s nimble, agile, and enables people to use data as an appropriately free flowing asset in the business and not a centralized bureaucratic process?”

My answer is that an effective business data standard is a pervasive part of the environment. Although a standard is in one sense a product, the utilization of a standard is an aspect of business processes – not a process in itself. A useful way to think of data standards is as a toolset you use to shape business processes.

Lubricating oil is a good analogy. Oil makes every machine in the factory run efficiently and extends the life of the plant. Different points in different machines need different oil types and regimes. But enacting these rules doesn't require a centralized Department of Lubrication. It's just part of the job.

Data Pipelines

Standards are about recognizing commonalities in order to remove barriers to sharing and exploiting information. When you spend time developing, promoting, or working with standards, it's not surprising when you become attuned to seeing parallels across domains. Challenges and opportunities encountered in other business areas resonate with you. Sometimes you can see how solutions in other domains address problems similar to those in your home domain. Other times, you catch neat articulations of your own concerns or conclusions, couched in interesting new ways.

Here's an example from a call for contributions to an oil and gas pipeline conference. Not my home turf. The organizers offer a number of prompts for submissions, including: “Construction Standards/Data Standards: What “should be/needs to be” captured when we put assets in the ground?”

It seems to me that working in the pipeline business has this major advantage for IT people: Everybody knows the business is about pipes, and they know pipes are laid in the ground. So issues relating to recording information about pipes (and other plant) are not far from the top of anyone's mind – no matter what part of the business they work in.

Insurance doesn't see itself as an infrastructure business. To us, a “pipeline” is an abstract concept. We'll think of sales pipelines, channels, workflows, connections, partnerships, networks. But most people in the business are naturally more oriented to products, contracts, rates, premiums and so on.

Insurance has always been about relationships, and with the growth of technology, relationships are increasingly enabled by systems. The industry has evolved into one where the data pipelines are crucial. The pipelines carry deals in various states of completion.

So perhaps we should be asking ourselves: What “should be/needs to be” captured when we put deals in the air? Okay, it's not very elegant. But it gets some of the flavor of today's – and tomorrow's – business, which is its circulatory, collaborative nature. This would help people to think about all the potential users of a piece of information, rather than the data requirements of a particular document produced at some point in the process.  Pipeline

Leading Standards

The Department of Transportation, led by CIO Richard McKinney, is showing true leadership with its commitment to creating data standards. Traffic and highways are getting smarter, and understanding traffic is increasingly important to all of us – drivers, transportation authorities, technologists, insurers, everybody. 

“McKinney, speaking at the SmartAmerica conference, said the transportation industry doesn't want to find itself 10 to 15 years from now with independently developed data standards that hinder communications.” Absolutely. If you want a vision for standards, take a look at the nightmare down the road – the mess you'll be in if you don't do standards.

McKinney said he doesn't know what role the DOT should be playing in standards creation, but he's clear about getting the topic addressed seriously across the industry. This sounds good – standards movements need a gameplayer, a connector, to get traction. Other government departments and agencies, please copy.   ComputerWorld

Smart City Standards

The British Standards Institution (BSI) has published a conceptual data model to help local government create smart cities.   This is a good way to bootstrap a standards process. By getting core domain concepts and relationships out into the community, interested parties can converse more meaningfully about priorities and details. BSI has worked with a bunch of local authorities and other organizations to create this starter model.   Smart Cities

Standards Heart of Health-as-a-Platform

Say what you like about official government publications, when they go for plain English they can make some striking statements. A new policy paper called Personalised health and care 2020: a framework for action from the UK Department of Health is very clear on the value of standards: “Standardization [of processes, datasets, platforms and interfaces] enables innovation, reduces development costs, lowers barriers to user adoption, speeds up wide-scale adoption and supports an almost infinite variety of bespoke and personalised service offerings.”

Digging farther into the text, I find that the policy aims to be robust on data standards – at least, that's my interpretation of the following: “At times, the [UK's] health and care system has tried highly centralised national procurements and implementations. When they have failed, due to a lack of local engagement and lack of sensitivity to local circumstances, we have veered to the opposite extreme of ‘letting a thousand flowers bloom’. The result has been systems that don’t talk to each other, and a failure to harness comprehensively the overall benefits that come from interoperable systems. In future, we intend to take a different approach. We will be tight on standards and definitions, and clear on expectations regarding interoperability, but we will support local decision-making on systems, programmes, interfaces and applications.”

It makes me smile to see the traditional evolution of disparate systems described in terms of flowers blooming. But the image maybe deflects from the implications of the message about future practice. Managing the dynamic between the implied centralization of data standards and local decision-making about technology will not be easy. The phrase “systems, programs, interfaces and applications” sounds like it might have been hammered out by a committee. It could be used to undermine “expectations regarding interoperability”. I wonder what “tight on standards” will mean in practical terms?

Personalized Healthcare

Customer Activated Ecosystem

IBM's Lynn Kesterson-Townes blogs compellingly on the need for insurers to have what she calls customer-activated ecosystems. One of her great diagnostics is: “How much time does it take to onboard an ecosystem partner? Do you ever have to say 'No' to an onboarding request because you can’t support the technical requirements?”

This two-part question should be asked in every organization – and repeated at regular intervals. It's only by pressing these points that real change will come about. We've got to accelerate away from the vague idea that it would be nice to be more agile, more open to partnership, more customer-centric, more real-time – and make it happen.    IBM

Future of Advertising Data

Where does advertising data go after tag management? Standards, of course. This is evolution: Industries sweat to cope with growing complexity until the point where they just can't avoid simplification. Standards will help advertisers and marketers understand data better, share data with each other, lower their costs, and speed up their processes.

Here's one of four things Josh Dreller says must happen in the industry: “Universal Data Standards. It’s still a bit of the wild west when it comes to data. Even if you can get your data easily out of your various customer engagement points, it’s hard to match up files from different systems. For example, one of the most common data fields in digital advertising is impressions. In some systems, it’s called impressions, but in others, it is called imps, served impressions, display impressions, viewed impressions, purchased impressions, etc. There’s literally a language problem between ad tech platforms that needs to get figured out before data can truly become as portable and flexible as it needs to be.”

But Dreller doesn't say who's going to bring order to the wild west of advertising data. They need the advertising equivalent of ACORD.   Marketing Land

Data Trumps Documents

All the best stories begin with “Once upon a time...” Here's a great one from the mortgage industry: “Once upon a time, documents themselves were considered the most critical element of the loan package… Now, the script has flipped, so to speak, so that data trumps docs in the eyes of investors and regulators alike.”

Once upon another time, the distinction between documents and data was only interesting to data processing nerds. Business folks' eyes would glaze over. Today, two drivers are making decision makers understand the deep value of this distinction. The first driver is regulation. The second is collaboration. As is often the case, these drivers also interact with each other.

If you need to bring the distinction alive for any colleagues who are not getting it, I suggest you try this. Documents are for printing, reading, distributing, filing. They don't provide information flow, because the information is trapped in them. Documents are coffins for information.  eLynx

Open Data

A new study says that open data could generate $13 trillion over the next five years in the G20 countries. It's notable that the study mentions the release of open data by private organizations as well as public bodies, which make up the usual providers of open data. The study also recommends countries make targets around open data.

I wonder if, eight or ten years from now, we won't be talking so much about open data, but about closed data. Think about it. If sharing data has such great value, then people will want data to be open by default. And since we all value our privacy, we will shift our attention to making sure that certain classes of data are not shared. The emphasis will switch from provision to protection.    Future Gov

The Swift Institute

SWIFT with its data standards and network is often viewed as the gold standard for industry data exchange.  I interviewed the head of Standards at SWIFT many years ago and found that they share many of the same challenges.

They created the SWIFT Institute a few years ago and I am impressed.  In fact, much of the work we are doing with the ACORD 20/20 effort may lead to my recommending something similar for ACORD.  The  "R" in ACORD represents research and driving industry innovation requires such an outreach effort.

Some may shy away from such collaboration as they think innovation is about competitive advantage. But when it comes to innovating anything involving data exchange with trading partners, suppliers and distribution channels in particular, the need for collaboration is imperative. And doing so frees up time to focus on innovations that do provide competitive advantage. Firms have more in common than differences.

The SWIFT Institute was founded in April 2012. Its core objectives are to:

• Extend understanding of current practice and future needs in global financial services, with a particular focus on transaction banking.
• Foster independent research by giving grants, access to research data and publication opportunities for approved projects.
• Act as a catalyst for bringing the SWIFT community and academics together to explore ideas, share knowledge and extend mutual understanding.

Here is a working document from their website and I include an excerpt of a familiar refrain.

Download SWIFT-Institute-Working-Paper-No-2014-005

The development of regional standards enables interoperability between banks, payment processors, and corporates in different countries. The development of a common language through standards can be a difficult process that requires centralized management, broad inclusion of stakeholders, and a coherent vision for payments integration. But developing a standard is not enough. Stakeholders need to implement a common standard in a uniform manner.

The ISO 20022 standards used in SEPA for credit transfers and direct debits were developed by the EPC with input from a wide range of European stakeholders. However, there have been variations in how banks and ACHs in different EU countries have implemented the SEPA standards, which has led to problems with interoperability. Now that migration to SEPA schemes is complete, regulators and other stakeholders in Europe can begin to modify the standards to ensure uniform implementation.

The task of creating deep enough rule sets that enable uniform implementation is difficult even in a domestic environment. When bringing together different domestic schemes under a regional scheme, it may be necessary to leave room for variation in the initial implementation of a regional standard. But too much variation can lead to a fragmented payments environment, which is incompatible with the principle of regional payments integration.

 The Swift Institute

Data Lock Out

Sharp focus from the legal profession on another reason why data standards are a good thing. Organizations are using the cloud more and more – although it's kind of worrying that “75% of CIOs had no idea how much cloud was being used in their companies”.

Lawyer Frank Jennings of the Cloud Industry Forum’s code of practice board says: “If there is an outage, you need to get back up and running as quickly as possible. It’s no use waving a contract around: you need a back-up plan.” And to get your data back and move it some place else, you need standards. Without standards, you can be locked out of your data, and unable to do business.   Gazette

Sharing Data

Big quote here, but I couldn't bear to cut it:

“One of One Mind’s open science principles involves adhering to widely accepted data standards. This makes sense because the standards help make the data useful. Sharing the data is not the end game. Using the data to accelerate the development of safe and effective treatments is what we’re after. Sharing data that cannot easily be interpreted or understood limits the value of the data. At the very least, it adds significantly to the cost and time needed to interpret and transform the data to make it useful. Using data standards adds to the value of the data in a measurable way, and promotes the positive effects targeted by data sharing and open science.”

I love the way blogger “shume” (modest, anonymous guy) puts the logic of standards together. “Sharing data” has become a shorthand for the benefits of data standards, but he rightly reminds us what the point of sharing is. We must always bring the discussion back to business benefit. In this case, shume is talking about data in the clinical research setting. But his comments apply to us all.   Data Sharing

Standards and Production Lines

Blogging on end-to-end processes in the clinical sciences, Dave Iberson-Hurst at Assero makes an interesting aside: “I await the howls of ‘science cannot be made into a production line’; I don’t ask that it is. I don’t want standards to drive the science, I want the recording of the science to be well structured using standards, that is a big difference.”

This resonates with business. Some people worry that bringing analysis and design discipline to business processes will clash with “the art of business”. But that's to mistake haphazard, evolved, or needlessly complex ways of doing things with creativity, innovation, or good service. In reality, you need efficient processes so you can perform effectively.

Standards are an important enabler in end-to-end processes. Standards help to connect islands of information and to create flow. They provide the conditions for creativity, innovation, and service to flourish – sustainably.

But as Dave says: “End-to-End is not just standards, its process, tools, standards and many other things working together to meet the business need.” Standards are not a panacea in themselves – but they are a constituent of every good solution.    Assero

Starting Standards with Concepts

Successful standards movements begin with innocent questions. Never assume that everybody shares the same understanding of even the most common terminology.

The American College of Obstetricians and Gynecologists (ACOG) is doing just that, beginning its data standards efforts by looking for conceptual consensus. They're asking questions like “When is a pregnancy full-term?” and “When is a labor considered spontaneous?”

From my experience with the standards process, this is the toughest part. It's the business part. There's ample scope for disagreement. It's also hard to manage, because these fundamental conceptual questions often spark other discussions.

However, these conversations also have a positive impact on the participants. They come to share other people's points of view. They realize sharing data means defining it in ways that make it meaningful and accessible to every relevant party.

Technologists are, if you'll pardon the term, midwives in this process. No one needs to mention any kind of computing system or communications network. This is all about the knowledge structures of the people involved, and the data they need to do their jobs.   Maternity

ACORD 2015 Advisory Committee Meeting

One of our primary goals this year is the ACORD2015 event in Boca Raton this November.  Our planning process includes the formation of an Advisory Committee to assist us in re-shaping and re-launching our flagship annual conference. I am truly excited about this new event and really appreciate the interest of our members. They came to our office today from near and far. The staff held breakout sessions on topics and we also did some videos with members of the group. We are looking to focus on innovation.  We realize that there are several audiences. Some need solutions today. Others want a peek into the future.  Lots of great ideas collected.

Applied Insurance Analytics - Here's the Book

I hear more and more about analytics today as I travel across the globe. Saporito says that most insurers fail to exploit their greatest asset, data. Because their use of analytics is largely limited to stovepiped operational areas such as underwriting, claims, marketing or risk management.  Applied Insurance Analytics demonstrates how to use analytics to systematically improve operations.  Even more important: it will help you drive more value everywhere by defining a focused enterprise-wide analytics strategy, and overcoming the challenges that stand in your way.

 

Saporito helps you assess your current analytics maturity, choose the new applications that offer the most value, and master best practices from throughout the industry and beyond. Throughout, she helps you gain more value from data assets, technologies and tools you've already invested in. You'll find new case studies, practical tools, and easy templates for improving the "Analytics IQ" of your entire enterprise.  I know Pat... a must read.   Get the book on Amazon

 

ACORD Webinar on Architecture (Listen!)

On January 21, 2015 ACORD’s Architecture team conducted a webinar (recording found here) on the mappings effort involving the ACORD Reference Architecture (Framework) and the ACORD Data Standards. This webinar was a continuation of a session that was held at the ACORD Implementation Forum this past October.  Shane McCullough is Chief Enterprise Architect. He can be contacted at [email protected].

We are pleased to announce a record number of registrants (262) and attendees (158). With such a large number of people showing interest, it is clear that ACORD’s continued efforts in this area are in high-demand. Because of this demand, we have set up a new Community (found in the Framework and Architecture Community Section on ACORD.ORG) to address member questions and to gather feedback. We also intend to host a follow up webinar in the coming months.

These efforts showcase the importance of the ACORD Reference Architecture in the future of ACORD. More and more organizations are looking to ACORD to provide guidance, tooling and services for internal projects as well as B2B data standards and forms. This webinar will provide a springboard into our ACORD 2015 event with multiple sessions dedicated to Architecture across the entire Insurance value chain.

Beyond BYOD

The spread of consumer technology is a much greater influence on people's behavior and preferences than their experience in the workplace. This is why the topic of BYOD (Bring Your Own Device) causes headaches for businesses. People want to use their own smartphones and tablets for work purposes – and why not? They don't want to break security, they just want the ease of use and flexibility that comes with using their own equipment. As our personal lives evolve in line with our use of technology, so we want to change the way we work.

Now that the good old “user” is actually the driver of IT, IT teams are working hard to reconcile the demand for flexibility with the needs of security and control. This is hard. But the bigger picture is all good. When you consider what this shift in leadership can do for our organizations, the benefits are huge.

First, and perhaps most trivial, is that support is shrinking. People teach themselves to use apps. They find apps themselves. They configure them themselves. They update them themselves.

Second, and more profound, is that ideas about exploiting information and connectivity are starting to come from ordinary people rather than appointed managers or visionaries. “Wouldn't it be cool if...” Everyone can now be a player in the world of collaborative IT.

Third, there's the impact on strategy. The generation coming through won't see data governance, and data standards, as an esoteric, technical issue. They'll say: “My car can't talk to my house – that's crazy!” and fix it. Better still, they'll say: “Wouldn't it be crazy if my car couldn't talk to my house? Well, how come my business systems can't talk to yours?” Likewise, they won't have any time for data lock-in – every cloud will have to enable data withdrawal, so standards will be essential. There will simply be no argument for non-standard approaches.

Stepping up for Standards in IoT - Things

Don DeLoach at Infobright says: “2015 will be the year when someone, whether an industry group or private company, steps up to establish the standards that will be accepted worldwide to ensure the proper technology is carried out to deliver on what the IoT promises.”

I hope so. But who? How do you get this party started? 

Check out the OMG pages as well.

Data Informed

OMG

China Embraces IMF Data Standards

Buried away in the details of the recent G20 meeting in Australia comes news that China has signed up to the Special Data Dissemination Standards (SDDS) of the International Monetary Fund (IMF).

Okay, so I can see why it didn't make the front page. Still, it's probably more significant than anything that did make the headlines. It's a sign that China is becoming more integrated in the global economic – and political – system. By progressing steadily with international standards, China is embracing transparency while at the same time gaining greater access to capital markets.

Events like the G20 are sometimes seen as an excuse for a party. Well, this time around, a great nation signalled its commitment to joining the world community for good. That's a cause for celebration.   China Daily

Cities as Systems - and Beyond

“In 2015, several Boston-area cities are planning clear data standards for consistency across different jurisdictions. That helps data scientists see cities as systems, rather than islands.”

What a great image. It applies to businesses too.

Data standards form a key part of the promise of open data. If open data doesn't add up, we're not going to get any meaningful results. The other challenge cited in this article is citizen engagement. In short, data can only be transformed into policy via the participation of the community.

Is there also a lesson for business here? It's common knowledge that data standards are the way to make data usable across the organization – to turn data into a real business asset. But the need to tap the community as the transformative power that turns data into real-world outcomes has become obscured, I think, by a misplaced faith in the autonomy of technology. That is: Some folks hope just having lots and lots of data will improve the business. Just like that.

To get value from data, you need to ask questions. So you need to have explicit goals that in turn suggest relevant questions.

To get value from data, you need to share it. So you need to understand who can make use of the data, and who can collaborate around it.

And to get value from data, you need to act on it. So you need to interpret your analyses, contextualize them, convert them into plans, and execute.

All of these activities are participative. They benefit from technology. But technology can't do the whole job. With data standards, you're set to make smarter decisions and lead the business in an informed and adaptive manner. Data standards liberate. They help us do what only people can do.   Beta Boston

HyperCat

The Internet of Things (IOT) has a new data standard. HyperCat is “an open catalogue specification that allows applications to discover and make sense of data automatically”. The new standard is aimed at ensuring data hubs don't become silos.

Without such standards, the IOT will surely die on the drawing board. The potential for using data streams created by devices in the environment is totally based on standardization. Standards are implicit in the whole idea.

Will HyperCat get traction? Well, they've got a logo; and they've got more than 40 organizations involved. Meanwhile, Google and Apple are doing their own things.

Whether the IOT evolves around community standards or de facto proprietary standards will depend initially, I think, on which type of technology component makes the commercial running. Will – for lack of a better term – end user devices be the driver of growth, or sensors?

Obviously, you've got to have both. But for me the sensors have the edge. Sensors can talk to each other, and to central systems, without needing to refer to, or be instructed by, any human peripheral.

For example, Fujitsu are growing (and selling, at a premium) lettuces using a clean-room environment, hydroponics, and sensors. Currently, they have people in contamination-type suits working the floor with tablets. But I guess that when they have enough data about growing the perfect lettuce, there won't be any need for the tablets. The sensors will be able to talk direct to the irrigation and lighting systems, coordinated by a central or shared rule base.

My guess, then, is that sensors and sensor hubs will be the faster growing part of the IOT by volume and downstream impact. Standardization at the sensor/hub side is likely to be established ahead of standardization via the end user device.

Of course, I don't know what Google and Apple are planning. Perhaps they're planning to go big into the sensor/hub side. Certainly, Google is interested in robotics and Apple are getting into control systems. It's one to watch.   FAST Company

Devil in the Details

True real-time business represents a big shift from business-as-usual. How well prepared is the business world?

I believe people are comfortable with the idea that long-period batch practices are incompatible with an always-on world. So they know they have to be faster. It's called near-real time. But I'm less sure we've all assimilated the idea of acting in real-real time. Perhaps we don't believe it's necessary.

So here's an interesting insight from predictive analytics in health. According to Justin Lanning of Xerox: “Many are actually using retrospective claims data to predict the future while others are using more recent data sources and others still are using real-time data.” And yet:

“If one EHR sends us a piece of data, we might know that the way they date and timestamp their data is different than this other EHR, and another one and another one. Maybe it’s stamped differently based on when someone opens a window as opposed to when they click ‘save.’ When you’ve got time-driven analytics, you have to know all that stuff, too. So it’s one thing to say that if we had real smooth interoperability, that all the data would come over, but it requires that attention to these little details you might not think about.”

In real-real time, seconds count. Think about it in the insurance context. In the not-very-distant future when your driverless car is involved in a collision with another driverless car, seconds will count. When the smart defenses in your office building react to an incoming storm, seconds will count. In both cases, fractions of an inch in positioning may count too.

Standards bodies must take real-real time on board and pursue its implications for business. We've always focused on details. Increasingly, those details are microscopic – but no less important.   Health IT