Skip to content


I Bet You Think This Song Is About You, Don’t You?

carly1

You’re so vain, you probably think this song is about you

You’re so vain, I’ll bet you think this song is about you

Don’t you? Don’t you?

You’re So Vain, Carly Simon, 1972

The song may not have been about you.

This post, however, is.

This post is not just about you, it’s about why so many things are going to be about you.

One of the reasons I speak at conferences is in order to learn. I learn by the comments people make. I learn by the questions people ask. The comments and questions help me refine my thoughts, adjust them, strengthen them, sometimes even discard them altogether. All that goes for Confused Of Calcutta as well: when I write a post, in a sense I’m doing the same thing I do when speaking at a conference. Expressing a set of ideas, explaining my rationale, responding to feedback and learning from that whole process.

For some years now I’ve been writing and speaking about a world where everything is connected, everything is capable of sensing, everything can publish, everything can subscribe. The Four Pillars theme I started this blog with is fundamentally based on publish/subscribe, and much of how I’ve viewed the enterprise world has been with pub/sub at its heart.

When everything is a node on the network, strange things can happen.

Take healthcare. I have had the privilege of watching people hard at work in hospitals a few times over the past two decades. Watching what they do. Watching how they do it. Watching the systems and processes they use, and the (often unstated) cultural values that bring it all together.

Last year, the reason for my extended stay at a medical facility was a relatively large tumour in my colon. Benign, but past its sell-by date. An urgency accentuated by the adjacent presence of a particularly vile attack of e.coli. So for nearly two weeks I found myself in San Francisco’s CPMC ; (salesforce.com, and Marc Benioff in person, ensured I had the best attention possible, something I will always be grateful for).

I couldn’t do very much. So I slept. And observed. And thought. And slept.

Lots of people rushing around. Some in serial patterns, some ad-hoc, some hard to describe. Machines and instruments everywhere. Some of them were static, some were wheeled around as needed, some were portable. Most had digital displays of some sort or other. Measuring my pulse and blood pressure. My oxygen intake. My temperature. Sometimes the measuring process was in stages: take samples of my blood and go somewhere with it, analyse it, come back with the results. Sometimes the instruments used were human beings, as specialists came and examined me at different angles and in different stages of undress.

Tons of information. Not just the things that were measured, but other things as well. Records of what medication I was given and when, what drips I was on and what the drips contained. X-rays and CT-scans. Records of the meals I had had, what I could have and what I couldn’t have. Records of my height and weight and even records of my bowel movements.

Tons of information. In tons of silos.

All brought together by someone writing it all out on paper, in notes that resemble a news feed. Some of it also brought together on the whiteboard on the wall facing the bed. Everything usually up-to-date, or at least close to accurate. Swivel-chair integration of the highest order.

Silos of information threaded together to create a composite view of the patient. Me.

These silos of information represent specialism; each specialist discipline comes with its own group of manufacturers and operators and providers and consultants. Heart stuff here, blood stuff there, insurance stuff in the other place. All brought together manually. With concomitant risks of error. Not always persistent; not always archivable; even if so, not always archived. Not searchable. Hard to retrieve. Even harder to share.

In some ways this was nobody’s fault. All these devices and instruments were built to do what they did, and nobody asked for their outputs to be shareable. Things were integrated, in their own peculiar way, but the integration took place around the discipline, the topic, the product. Heart things worked with other heart things. Blood things worked with other blood things.

Now, when everything’s a node on the network, we can break these silos. Take each sensor that relates to a particular person, and build a composite view of the person rather than of the discipline or product.

That’s one of the key benefits of the internet of everything, the ability to aggregate feeds around concepts that were hard to aggregate around before.

Concepts like patients. And actually being able to see all information to do with a patient in one place. The 360 view of the patient. Accurate and up to date. Persistent, archivable, retrievable, searchable. Shareable.

Concepts like passengers. Information in the airline industry was often organised around their product, the flight. Passengers were not as important as flights. A passenger was associated with a flight. And her luggage was also associated with a flight. Change the flight and the risks of passenger and luggage travelling together were reduced…. because often their association was nominal except at the flight level.

Concepts like drivers. Concepts like citizens.

Concepts like customers.

Getting 360-degree views of patients, passengers, citizens, drivers, students, customers. Taking the siloed strands of information out of the siloes, strand by strand, and rebuilding them around the person.

You don’t have to be vain any more to think that this song is about you.

Because it is. Because, for the first time, it can be.

Posted in Four pillars .


14 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

  1. Jeff Mowatt (@HomelessDead) says

    Shifting the perspective from product centered to people-centered has been my aim JP. Part of that involves breaking down silos.

    The experience of being treated for Leukeamia gave opportunity to compare our NHS to the US system when India took a stnd against Novartis.

    http://www.p-ced.com/1/node/75

  2. Philly says

    Reminds me of the epic effort it took to get our happy folk all those years ago in just one institution to centralise common data… holy hell. I think the biggest obstacle to the global profile is the antiquated mindset of “knowledge is power” and the inherent almost sociopathic anxiety that data sharing drums in those keepers of the gate of said data. George Orwell did us no favours in that regard.

  3. Rathi Krishnan says

    Yes, through the internet a plethora of information is collected and stored and it’s wonderful to have all this information so readily accessible and comprehensible. However, there still needs to be a human willing to analyze the data, make valuable connections, pass judgment, and make decisions. In older times, a medicine man or a healer would be responsible for entirely healing a person, yet in a modern world every whisper, every whimper is broken down to discrete elements, making room for more error and misjudgement. Unfortunately, the modern man’s needs and lifestyle are largely responsible for this far more complicated, fragmented, and not necessarily more satisfactory existence.

  4. clive boulton says

    Very inspiring. Leslie Wu PhD thesis got into the Swivel-chair integration use case.
    https://www.facebook.com/events/144468582396835/

    But I think we need more graph databases. Traditional databases are blocking nodes in the enterprise. Learning more at GraphLab workshop. http://glw2.eventbrite.com/

  5. Rob Paterson says

    So True JP Thanks Rob

  6. John Taysom says

    A wonderfully crafted post JP. And a great song….but specialisms are another thing too. They are concentrations of ontologies where words have specific contextual meaning. So far we have found it hard to enable machines to spot the context switch and hence to change the connections. I remember vividly my biology teacher explaining the beauty of our own pub/sub system by noting that when you put on your jacket you know it’s on. But you don’t need to know its on all the time so the subs turn down the signal. Others become more important. The idea of signal strength in context and balancing signal strengths one against another so the appropriate signal is acted on need to be layered onto pub/sub. I wonder how these will be dealt with in a world without specialisms? Ontologies and issues around id (as disabiguated instances of “me” come to mind. What about privacy issues? And who/what conducts the performance?

  7. JP says

    @Philly yes, that’s why we can look forward to some magic here

  8. JP says

    @rathi I guess I remain optimistic that we are evolving to a place where we can look at ourselves more holistically again

  9. JP says

    @clive thanks

  10. JP says

    @rob thanks

  11. JP says

    @john I don’t decry specialism. I use it to explain why we had the silos in the first place, and why we have a need to bring multiple specialisms together when we look at the human being from the viewpoint of health or welfare or education or even business

  12. JP says

    @john see if my next post on the subject helps, the Carole King one

Continuing the Discussion

  1. You Just Call Out My Name: Continuing with the internet of everything – confused of calcutta linked to this post on June 30, 2013

    […] This is a follow-up to my post yesterday on the same […]

  2. Healthcare on the brink of a digital disruption | SHiFT linked to this post on September 30, 2013

    […] well connected machine. Our healthcare systems and our patient data not so much as JP Rangaswami recently wrote based on his personal […]



Some HTML is OK

or, reply to this post via trackback.



%d bloggers like this: