Vectors

On June 1, 2007, I was walking up Massachusetts Avenue from Harvard Square, when I noticed a blue flash on the periphery of vision in my left eye. It showed up whenever I moved my eyes rapidly, or took a bouncing step. This didn’t disturb me until I looked at the sky and saw that my left eye had become a snow globe. There were more floaters than I had ever seen before, some in focus (that is, against the retina) and some blurry (farther from the retina). I assumed this development related to the continuing blue flashes and immediately worried that my retina might be detaching.

So I went to the Harvard urgent care facility, where the doctor on duty promptly sent me by taxi to Massachusetts Eye and Ear, where, after much waiting around, the ophthalmologist told me I was witnessing something called posterior vitreous detachment, or PVD. It happens to most of us as we get older. And, while it carries a risk of retinal detachment, that wasn’t happening in my case. He told me I shouldn’t be worried or alarmed, when the same thing would happen to my right eye. (Which, a few months later, it did.)

When I got home after all this, I looked at the accumulated paperwork: from Harvard University Health Services (HUHS), from the taxi driver, and from Mass. Eye and Ear. All three had spelled my surname wrong, and in three different ways. There were other errors scattered around, including my birth date, my employer, and so on. To my knowledge, nothing in the medical information was wrong, but, not being a doctor, I couldn’t be sure.

My experience through the next three years as an HUHS patient was a good one. The care was excellent and, in one case, saved my life. But there were also lots of glitches that had everything to do with poor communication of data (especially between hospitals, radiology labs, and various specialty centers).

That story has a happy ending because I lived, and I’m still doing fine. My mother wasn’t so lucky. She died following a stroke that probably would not have happened if two of the medical departments attending her had communicated fully about the medication she required after a surgical error during a routine gallstone-removal procedure. The surgical error was a risk she accepted when she signed the release for the procedure. But the pharmaceutical error was a different matter. Had she been taking the proper anticoagulants, she might not have had the stroke.

Ours is not a litigious family, and at ninety, Mom had lived a long and very full life. But bad data is what killed her, and it’s unlikely that big data could have saved her.

What both these cases call for is a combination of holy grails for medical system reformers. EMR (electronic medical records), EHR (electronic health records), and PHR (personal health records) are all intended to reduce errors and guesswork by making medical data easy to share between and among individuals and various health-care providers. All help make the patient what Joe Andrieu calls the point of integration for his or her own data, as well as the point of origination for what gets done with it: “When we put the user at the center, and make them the point of integration, the entire system becomes simpler, more robust, more scalable, and more useful.”³

In respect to the story about my eye problem, Joe asked,

But what if those systems were replaced with a VRM approach? What if instead of individual, isolated IT departments and infrastructure, Doc, the user was the integrating agent in the system? That would not only assure that Doc had control over the propagation of his medical history, it would assure all of the service providers in the loop that, in fact, they had access to all of Doc’s medical history. All of his medications. All of his allergies. All of his past surgeries or treatments … All of these things could affect the judgment of the medical professionals charged with his care. And yet, trying to integrate all of those systems from the top down is not only a nightmare, it is a nightmare that apparently continues to fail despite massive federal efforts to re-invent medical care.⁴

EMR, EHR, and PHR are VRM solutions, meaning they match up with the lists of VRM purposes and tools I shared earlier. While my own leverage is on projects that will show results in less time, I want to honor the good continuing work of Adrian Gropper, Brian Behlendorf, Jon Lebkowski, and others in the VRM community working on health-care solutions. Many others like them have been on the VRM-in-health-care case since long before I showed up, and many more will still be working on it after I’m dead. I can’t think of a bigger, tougher, or more important challenge, especially in the United States, where health care is an unholy mess.

-Driven versus -Centric

Joe’s post, just cited, was highly clarifying for everybody working on VRM tools. But we still had a problem with “user-centric,” because the point of view seemed to be anchored outside the user. Exactly who was being user-centric? Was it the user, or some second or third party?

Then, on April 21, 2008, VRM developer Adriana Lukas put up a post titled, “Two tales of user-centricities.” She wrote,

Last year at the IIW in Mountain View, I got talking to Bob Frankston about the difference I started to see between the user-centric and user-driven. Bob, in his inimitable fashion, used the tuna salad we were having for lunch during the conversation to coin an analogy. A ready-made tuna salad is user-centric—it has been decided what goes into it, in what proportions and what order. It has been designed around me and for me but I cannot add anything to it.

Giving me ingredients, utensils and a recipe suggestion and letting me get on with it, leads to user-driven design—it can still be meant to become a tuna salad but I get to put it together, determine the proportions, skip or add ingredients. The process is driven by me …

Of course, there are times for user-centric and there are times for user-driven. Not everyone wants to make everything themselves and neither is it the best or most effective way to design all systems or tools. But there are cases when only user-driven will do. And VRM is one of them.⁵

The difference is one of perspective: “-centric” is anchored outside the individual, while “-driven” is anchored inside the individual. The difference is also one of intent. Am I doing the intending, or is somebody else intending for me? (I still bristle every time some company puts words in my mouth, saying “my” on my behalf. For example, MySpace was never anybody’s space but the site owner’s.)

One year later, as we continued to weigh the differences between user-centric and user-driven, Joe put up a series of blog posts on user driven services. (He leaves out the hyphen.) In it, he provides fresh guidance to both the customer and vendor sides of the marketplace:

User Driven Services put users in charge. Users start each interaction, manage the flow of the experience, and control what and how data is captured, used and propagated. Users are the cause and the controller, working with service providers to co-create collaborations that create value for all parties.

From self-serve gas stations and soda fountains to ATMs and self-checkout grocery stores, companies have been putting users in charge of different aspects of their services for years. With GetSatisfaction—which allows users to self-organize for cooperative customer support—and Facebook—which provides social context for user-generated content—users are not just self-servicing, they provide the core content behind the user experience. Now, through user-centric Identity and API access to most popular online services (Flickr, Twitter, Facebook, etc.), users can direct which parts of their experience are serviced by which providers, allowing unprecedented realtime flexibility in service creation.⁶

Joe then rolled out ten blog posts explaining the virtues of user-driven services. They include “Impulse from the User,” “Data Portability,” “Service Endpoint Portability,” User Generativity,” and “Self-managed Identity,” among others. But the verb impelling them all was drive. His encompassing message: customers are not just going to be passengers anymore. They are going to be drivers—of themselves and of others helping on the their side of the demand-supply relationship.

Yet, there was still some ambiguity.

In June 2011, in a discussion thread on the VRM list titled, “VRM tool characteristics” (a thread that helped draft the list in chapter 17), Joe noted that ambiguity and wrote, “I expect that the idea that [the]customer is the recipient of value is so inherent to the entire conversation, that we just don’t think about it. Like a fish who doesn’t know about the water … but if you think about it, corporate software does NOT have this mandate. Corporate software is there to create value for the corporation. VRM is there to create value for the individual … There is something vital in the idea of tools that allow the individual to create value for themselves.”⁷

Possessive Pronoun Mash-up

Following the exchanges above, Iain Henderson posted, “The Personal Data Ecosystem” at the Kantara Initiative’s site.⁸ In it, he sorted data into four categories, in four overlapping circles:

1. My data (me)—Stuff only I know or should have
2. Your data (service providers, governments, retailers)—Stuff they use to remember me, or have gathered about me
3. Their data (marketing data providers, credit bureaus, search services)—Stuff they’ve found out about me, somehow
4. Everybody’s data (public domain)—Stuff everybody can find out

Iain expects that “over time some 80% of customer management processes will be driven from ‘My Data’,” and gives two reasons: “(a) because we are already seeing the beginning of the change in the current rush for ‘user generated content’ … and (b) because the economics will stack up.”⁹ Those are the economics both of intent (customers expressing demand directly) and of cost-benefit. “Organizations care less about the sources of data than its usefulness.”

It is no coincidence that the U.K. government in 2011 launched the “Midata” initiative.¹⁰ Iain and his colleagues, along with many others in the U.K., had long been pressing the government to take a policy stand recognizing citizens as responsible parties for data collected from (and about) them. In the last national election, all three major political parties favored that position (compared to none in the United States). The winning coalition government followed up with Midata, a voluntary program the government is pursuing with industry to give individuals the ability to download their personal data in a portable electronic format.¹¹ The aim, says the government, is that individuals “will be able to use this data to gain insights into their own behaviour, make more informed choices about products and services, and manage their lives more efficiently.”¹²

In April 2011, the U.K. Cabinet Office Behavioural Insights Team (of the Department for Business, Innovation and Skills) published, “Better Choices, Better Deals: Consumers Powering Growth.” It says the office’s effort “does not set out a new legislative programme. Nor will it see a range of new regulations laid in Parliament. Rather, it seeks to put in place a wide range of new programmes that have been developed in partnership with businesses, consumer groups and regulators.”¹³

I met with three members of the U.K. team in August 2011, and sensed that what they’re doing is studiously light-handed. From the same document:

We will see two profound changes:

• A shift away from a world in which certain businesses tightly control the information they hold about consumers, towards one in which individuals, acting alone or in groups, can use their data or feedback for their own or mutual benefit.
• A shift away from seeing regulation as what Government-sponsored bodies do after consumers have suffered in some way, towards one in which individuals and groups feel more able to send the right signals to business, and hence secure the products and services they want.

In short, we want to see confident, empowered consumers able to make the right choices for themselves—to get the best deals, demand better products or services, and be able to resolve problems when things go wrong.¹⁴

Midata is citizen-centric rather than citizen-driven at this stage, but it’s looking for more of the latter than the former. More importantly, it states that expectation.

A couple years back, over beer at a pub somewhere in London, Iain explained to me why the U.K. was at the VRM forefront. “A stool needs three legs,” he said. “The three in this case are business, the customer, and government. Having the government support both business and the customer is a big advantage.”

U.K. members of the VRM community, including Iain, William Heath, and Alan Mitchell, have been on the case from the start. In November 2011, Alan told me “becoming VRM-ready” has been central to their argument to business. For the rest of the argument, he gave this list:

• Improved data accuracy and quality
• Reduced guesswork and waste
• New insights into customer needs and behaviors
• The chance to innovate new services

And it appears to be working. As of that same month, Callcredit (which keeps files on every adult in the U.K.), Scottish Power, and the Royal Bank of Scotland have all publicly lined up behind Midata.¹⁵

Personal Data Stores/Lockers/Vaults/Clouds

As of this writing, Mydex (the company Iain, Alan, and William cofounded), Azigo, Personal.com, Privowny, Qiy, and Singly are all working on PDSes (Personal Data Stores), also known variously as personal data lockers, vaults, and clouds. All are points of integration for one’s personal data. All are described in different ways, which will change by the time you read this. Behind some of these are noncommercial and open source development projects, including the Locker Project, KRL, Pegasus, Project Danube, TeleHash, and WebFinger. Many open standards are also involved, including RSS, Atom, Activity Streams, JSon, evented APIs, and others.

Here is how Jeremie Miller put the challenge in a speech to the Web 2.0 Summit in November 2011:

You need to have a home for your data. I’m trying aggressively to define this home, in … the best software, the best technology, the best legal terms. This home is yours—that you own, that you control. And this home is for your data.

This ability for you to have it and share it out, is going to transform our industry, over the next ten years. There is going to be this tectonic shift, as everything sort of re-shapes and re-centers itself around people, around individuals, and around the mountains of data that they have … Everybody talks about “big data.” This isn’t big data. This is going to be the era of small data, of my data.¹⁶

Let that era begin.