Sharing Test Results in the Same Way

My Photo
Name:
Location: Plainfield, NH, United States

Thursday, June 16, 2016

Seeing the Forest for the Trees: Understanding the YouBase, UnPatient and Healthbank Transatlantic Collaboration



I). YouBase: http://youbase.io (a peer-to-peer data store and personal information exchange trust layer for a more personal health science)

II). UnPatient: http://unpatient.org (a non-profit, social benefit organization)

III). HealthIT TV: Debating the Utility of the Blockchain in Healthcare (Leonard Kish, Michael Dillhyon, Dan Munro: Apr 11, 2016. 25:34 min): https://www.youtube.com/watch?v=LEgljmCA9I0

################################################################################
IV). https://healthbank.coop (6-11-16)
https://healthbank.coop/newsletter-archive   (June 2015 to April 2016)

Healthbank receives Global Digital Health 100 Award, as one of the world’s 100 most innovative digital healthcare companies (12-10-14)


The Swiss cooperative healthbank has been honored as one of the top 100 digital healthcare companies in the first Global Digital Health 100 Awards 2014, by The Journal of mHealth.

Reflecting the evolving importance of mobile technologies and digital solutions in the healthcare sector, the Global Digital Health 100 acknowledges the significant work being carried out by pioneering organizations from around the globe. “This year has been amazing for digital health” said Matthew Driver, Managing Editor of the Journal of mHealth. “There are so many great companies producing really innovative and amazing products that are set to revolutionize healthcare delivery. We had a very difficult time narrowing the pool and selecting the final 100. We know that all the 2014 honorees will grow into some amazing companies that are sure to make an impact.”

The Award represents 6 months of research by the editorial team at the Journal, who considered the offerings and innovations from over 1,500 different mobile and digital health companies worldwide, before selecting the final 100. healthbank CEO Reto Schegg commented: “healthbank is honored to be recognized as a top 100 global digital health company, and to be included with so many highly innovative companies in this fast-growing sector.”

The full list of honorees can be found at www.thejournalofmhealth.com *

################################################################################
V). Healthbank (A citizen-owned global health data transaction platform)

Michael Dillhyon

Healthbank Founder & Member of the Board

Biography:

Michael is the founder of healthbank (in 2012), and served as CEO until July 2014. In 2013, he was the first Entrepreneur-in-Residence for SystemsX.ch, Switzerland’s largest early stage life sciences fund, and served as a mentor for the ETH Innovation and Entrepreneurship Lab. Previously he served as Chairman of Genebio, a bioinformatics software firm, and sat on the Strategic Advisory Committee for HealthTIES, an EU-backed consortium of four of Europe’s top biosciences regions. Previous to his move to Switzerland, Michael co-founded two US-based firms: Netelligent Corporation and ActiveObjects, where he held the roles of President, Chairman and CEO respectively. 
https://www.linkedin.com/in/healthbank (Michael Dillhyon LinkedIN Profile)
https://www.healthbank.coop/team/  (Operational Team)
https://www.healthbank.coop/team/#BOD (Board of Directors)
https://www.healthbank.coop/team/#advisors (Advisory Board)

############################################################################

Unpatients—why patients should own their medical data

Leonard J Kish & Eric J Topol  .                              
Published online (08 September 2015): Nature Biotechnology 33, 921–924 (2015) doi:10.1038/nbt.3340  

For the benefits of digital medicine to be fully realized, we need not only to find a shared home for personal health data but also to give individuals the right to own them.

It's often said that data are the new gold, or the new oil, but they are much more like a New World distinguished, at least in part, by new maps. Indeed, the planet is becoming a new world of relationships, descriptive data and information flows. There are now over 1.5 billion registrants on Facebook (Menlo Park, CA, USA), and a Swedish startup called Truecaller (Stockholm) has assembled a phone directory of >1.6 billion human beings, with the intent of having every person on the planet in its directory. Social graphs that depict relationships between people and organizations are the new maps of a connected humanity—maps of people, organizations and many other dimensions of data that reveal how things are related. As recent examples, we've seen months of activity data from 22 million Americans and over 250 million nights of sleep data1, 2. Such global data efforts have not yet reached medicine, but their arrival is both inevitable and imminent.

In parallel to these social graphs and global data sets, there is an unprecedented and rapidly developing capability to digitize a human being. Creating the equivalent of a Google medical map or the medical essence of an individual would integrate multiple layers of phenomic, physiologic, anatomic, biologic and environmental information3. Just about everything that makes a human tick can now be quantified like never before, by means of sensors, sequencing, laboratory tests and scans. Recently, it has been shown that a single drop of blood could be used to reveal the virome of an individual's exposure, uncovering not only which viruses the person was exposed to but when4, 5, for just $25. This exemplifies our newfound and accelerated ability to capture and analyze human data, which most of us could not even fathom a few years ago.

Such medically relevant data from an individual is not a one-off gathering. Rather than simply falling under the definition of 'big data', the data can be, and often are, obtained longitudinally, over the course of a lifetime, fulfilling the idea of 'long data'. Furthermore, such data are contextualized, often in real time in a person's real world. Enabled by mobile technology, an external 'wisdom of the body' (in contrast to Walter Cannon's classically described autoregulation, homeostasis, in his book Wisdom of the Body6) can be developed, with feedback of integrated data to the individual (Fig. 1). Soon enough, virtual medical assistants will emerge that incorporate machine learning about a person, and could include everything medical, as well as the person's lifestyle, behavior, social network, finances and how they are interrelated. Quickly, one can imagine that, just from a watch that collects blood pressure with every heartbeat, terabytes of data can and will be generated on an individual basis. Much of the data will fall into the category of patient-generated data and will ultimately eclipse the amount of data captured today in clinical electronic medical records.

Figure 1: The medical data ownership engine. (See original article)

Each individual gets direct feedback of her/his own generated data through biosensors, physical examination tools, laboratory and imaging tests, comprising a new 'external' wisdom of the body. Such data are fed into the flywheel of the engine and eventually, when there are enough individuals amassed into a big medicine resource, there is a breakthrough to form a valuable medical knowledge resource. That, too, provides external feedback to the individual for optimal prevention and medical treatment.

 “This is a unique moment where we may be able to provide for personal control and, at the same time, create a global knowledge medical resource.”
Yet currently there is no 'home' for such data over time, at either the individual or the population level. Although there are early proposals for how some of it could be bundled with one's electronic medical record7, it seems unlikely this will occur, in the United States at least, given the landscape of balkanized health records and multiple providers of care for each person. Ironically, we're looking at the prospect of a new, high-definition picture of individual human beings, and at the same time for that person's data to be homeless, dispersed and inaccessible. Where the data live will determine the maps we can create and the directions we can go in with health, both as individuals and as a society. We propose here that the key step to liberating personal health data and realizing their true potential in human research and clinical practice is the provision of data management systems that give individuals the right to own their own data. The technological advances developed for evolving digital currency systems, which allow individuals to hold and secure digital assets without a central authority, are being used to create new digital property systems, including personal medical data property. Whatever the means, it is critical for individuals to seize ownership of their data in order for the real benefits of a new, data-driven high-definition era of medicine to be actualized.

Data, data, everywhere and nowhere:

Today, in the United States, health data live in a plethora of places, from electronic health record (EHR) systems, insurance claims databases, siloed personal health apps, research and clinical trial databases, imaging files and lots of paper. Although seemingly everywhere, any true semblance of an overarching organization or standardization of medical data are lacking, whether at the individual or societal level. Health data are, categorically, quite difficult to move from one place to another, and there are few to no incentives for sharing, a situation that leads to extensive data hoarding. In the United States, despite $30 billion in incentives to get data flowing, the problem of health data locked into proprietary EHR systems is so bad that the US Office of the National Coordinator for Health IT recently released a report on 'information blocking', a term unique to the US healthcare system8.

Even worse than not being able to get to the data, disorganization and balkanization contribute to poor outcomes and death. According the US Department of Health and Human Services (Washington, DC, USA), an estimated 20% of preventable medical errors are due to the lack of immediate access to health information9. Of the estimated 400,000 preventable medical errors leading to death in the United States annually, we can project that 80,000 people die every year (or 220 per day) because of the lack of needed access to medical information.

Furthermore, the US legal framework is constructed in a manner to block individuals from accessing their own medical datain 49 of the 50 states in America, these data are owned by doctors and hospitals10. This ownership model is an outgrowth of an entrenched paternalism: the medical community's belief that patients are unable to handle or deal with their data. Rigorous studies have proven just the opposite11; patients are fully capable of possessing and managing their own data, a capability that not only increases their sense of well-being but also enhances bonding with their physician12. Meanwhile, the Health Insurance Portability and Accountability Act (HIPAA), defined in the pre-internet era, has largely become an excuse not to share with anyone, including patients. It's time to recognize we are in a new era, where patients have the tools and incentives to do much more.

In contrast to the legal and technical difficulty an individual faces to obtain all his or her own medical data is the relative ease with which hackers have managed to breach ~100 million patient records in the first half of 2015 (ref. 13; Fig. 2). The larger the set of data, the more attractive it becomes as a target for hacking. On the other hand, not even 10% of Americans have attempted to access their EHRs, let alone found that information in a format that was interpretable14. Clearly, a new solution is needed.

Figure 2: The timeline for electronic medical data hacks in the United States of over 1 million individuals. (See orignal article)

The graphic does not include a large number of hacks in this time period <1 million individuals.

The need for outright data ownership:

In no other walk of life does an individual pay for a service but not own what they have bought. Yet in medicine this has been tolerated for as long as there has been a paper record, dating back hundreds of years, despite the fact that the patient has a vested interest in the uses of that information—it is he or she whose health is on the line. In the coming years, for many individuals, this issue will be further exacerbated—massive patient-generated data sets will be flowing through their mobile devices, which they own. At present, no system for integrating such data between apps exists. Anticipating this important unmet need, we feel the time is ripe to develop such a platform that brings together all of an individual's data
·       traditional doctor-ordered laboratory tests, scans,
·       visits and
·       patient-generated data.

Historically, the concepts of property and democratization—spreading decision-making and knowledge—are deeply intertwined. Property and happiness were inextricably linked in 1776; the US Declaration of Independence's declaration of our rights to “life, liberty and the pursuit of happiness” is thought to derive from John Locke's writings on the right to “life, liberty and property.” And property rights were a driving force of the revolution. A patient-centered medical revolution will also require new rights. A platform that captures all of a patient's data cannot simply be accessible to patients; it needs to be owned and therefore controlled by the individuals who contribute to it.

Yet a common refrain among governmental healthcare leaders is, “it's not about ownership of data, it's about access and control”15. Such a construct also overlooks one of the oldest rules of law. The phrase “possession is nine-tenths of the law” arose because it is relatively easy to enforce ownership rights if one has possession of something, but difficult to enforce otherwise. The same is true with data. Simply put, if you don't have possession, you will have to ask permission and be granted access. A person cannot maintain true control when data live on someone else's server. A recent article entitled “The future of the web looks a lot like bitcoin” perhaps summarized it best: “We don't own our data; we just visit it from time to time16.

Let's not underestimate the economic value of ownership. For traditional (nondigital) forms of property, clear ownership enabled by networks of trust has been argued to be the greatest difference between first world and developing world economies. Property drives the flow of commerce. Transactions become trusted when ownership is clear and networks are created to link seekers and sellers of assets, leading to a greater number of transactions and greater transparency. According to economist Hernando de Soto, “The moment Westerners were able to focus on the title of a house and not just the house itself, they achieved a huge advantage over the rest of humanity.” The data on the ownership of the house, and a network to enable transfer of and use of assets, creates more flow of transactions and better economics. To create a health data economy, we need to provide the same trust and increased flow.

We continue to see the same principles play out in the most successful platforms of this era of the internet. Similar mechanisms are playing out with Uber (San Francisco, CA, USA) and Airbnb (San Francisco), now multibillion companies that have simply enabled platforms for what are initially 'surplus resources' of car transportation and shelter, respectively, and now have created economies of their own. In the case of Airbnb, some areas are being revitalized with tourists in towns where sleeping accommodations were previously inadequate. An emergent system more responsive to individual needs has arisen.

And we suggest the same can and will be true with health data. It's not so much about the data; it's about the ownership of the data. Without ownership, there can be no trusted exchange. What we have now is like trying to create an Uber without knowing who owns the cars or an Airbnb without an owner of the homes. It's common to hear how our health system should be 'patient-centered', but without clear rights around how it can be used, shared and leveraged by the individual, it cannot reach its full potential. To build a truly thriving health data economy, we need to harness the power of data ownership.

Of course, data are different from physical assets. Data can be easily copied yet become more, not less, valuable when shared, aggregated and analyzed. But like physical assets, medical data sharing can be likened to the 'tragedy of the commons', where individuals acting independently and rationally according to each's self-interest behave contrary to the best interests of the whole group by depleting or hoarding some common resource. Still, some of the same mechanisms can be used to enable the exchange of health data in the same way platforms and ownership enable the exchange of real world assets.
Health data are certainly another kind of 'surplus resource'; it's not currently creating much value without ownership and a platform to facilitate exchange. Yet if such a system were enabled, it would allow sharing and transactions of data globally in a more adaptive fashion. Each party would have an incentive to join as each could make their own proprietary data more valuable by contributing and receiving other's contributions.

Patients are already motivated to contribute. Polls consistently show 80% of consumers are eager to share their medical information, provided its privacy and security can be assured17. The ability to share the rich granular data from each individual to create a global resource leads to a flywheel effect (Fig. 1). Two feedback loops are engendered by the medical data ownership engine: the external wisdom of the individual's body and the wisdom of the population participants. Data matching the 'nearest neighbor' Google medical map from the population to the individual provides a potentially valuable knowledge resource for improved prevention and treatment—a resource that heretofore has not existed but is now eminently attainable. If we can amalgamate more than a billion members for a social network or a global phone directory, surely this, too, can be achieved—if we can override the obstacles.

Are personal data, in particular biological data, forms of property to be bought and sold? Should they be? Since the late twentieth century prominent legal scholars have been mostly divided into two camps regarding 'propertization' of personal data:
·       entirely for it, mostly from a techno-libertarian perspective or
·       entirely against it, fearing for privacy and human dignity.

Finding common ground between these two perspectives has been historically challenging. One reason, we submit, is because the solutions have not been in place to easily facilitate both trade and privacy. Through the efforts of a nonprofit, social benefit organization (http://unpatient.org/), we have proposed a technological solution:
·       that allows biomedical data to be shared and traded as property at a very granular level, but that
·       retains the privacy and security necessary for human dignity and in compliance with existing regulations.

For health data to have a real home owned by its rightful owner, they need to be:
·       first, accessible anywhere and always available to the originator;
·       second, controlled by the person they came from or their agent;
·       third, unique and verifiable as belonging to a real person;
·       fourth, privacy-enabled;
·       fifth, secure;
·       sixth, independent of any third party; and
·       finally, able to solve the data provenance problem, that is, when, where and from whom the data came.

An ideal digital ownership system would:
·       foster better trust in the accuracy of data;
·       connect people to facilitate enhanced sharing, anonymity and security;
·       create a single system of exchange,
·       standard methods of exchange and
·       better metadata to assess the value of a piece of information; and
·       finally, enable ways for all involved to benefit from sharing so as to maximize sharing and value.

Where to go from here:

How could we transition to a new model and construct such a system? Bitcoin's underlying technology points to ways to enable digital property on a global platform. Bitcoins are bits of digital property ('coins' are a ledger of transactions, a shared database of who owns what at a given point in time). Ownership is enabled by network consensus. Although there are certainly negative impressions about bitcoin, its digital ownership model creates a shared, agreed-upon record of data. Using and repurposing of the block chain, wallets and 'proof of work' components are already being adapted for multiple forms of data that are beginning to look like a global, distributed data ownership store. A worldwide health data graph enabled by health data ownership may not be far behind16.*

*Peck, M.E. The future of the web looks a lot like bitcoin. Spectrum IEEE (1 July; accessed 2 July 2015). http://spectrum.ieee.org/computing/networks/the-future-of-the-web-looks-a-lot-like-bitcoin

Once the infrastructure is built with clients and nodes for such a data network, transferring secure health data could be as easy as sending an e-mail is today. When an individual wants to receive an element of their record, the data sender sends it to the individual's public address or public key (which could be displayed as a QR code on a smart phone). It would then be signed by the sender's private key and could be opened only with the individual's private key.

Different health data structures could be accommodated within different wallet address structures. The creation of a new global infrastructure for data, accessible through Bitcoin wallet addresses, or something like them, has the potential to provide a universal patient identification mechanism, which separates personal information from health data as each data element can exist and be trusted independently. Peer-to-peer data stores could provide the mechanism to save any kind of file type independent of a third party, much like Bitcoin is intended to provide stores of value independent of any central authority. Peer-to-peer networks of data would also offer better adaptation and responsiveness to changing conditions, creating a system that evolves and learns quickly. In addition, block chain–related technologies could help solve the data-provenance problem as such systems would provide a record of when and from where a piece of data came, with a digital signature, creating the potential to visualize a data supply chain.

There are already models to achieve individual ownership such as Switzerland's Healthbank, an entity owned and governed by its citizen members (https://www.healthbank.coop). For no charge, Healthbank empowers users to store, manage, share and benefit from their personal health information according to each user's individual needs. It has the intent to create a global data transaction platform to support medical research. The benefit of using a distributed, peer-to-peer data store, however, is that it would not need to be managed or controlled by any third party, no matter how well intentioned; it would be controlled by contributors on a truly global scale.

A global resource for hosting and sharing personal health data would have compelling benefits not only for medical research and treatment but also for current data hoarders if they choose to adapt. Imagine if there were minimal storage fees and data management fees, but individual records were a part of the universal store. Costs for each participant would be minimized greatly, whereas each element of data, because of the flywheel of virtuous feedback (Fig. 1), would become more valuable, and many interoperability problems would no longer belong to the provider and their vendor. Without supporting and contributing, current providers and vendors could become the taxi companies and hoteliers caught fighting for survival as new, more effective platforms begin to compete.

Conclusions:

We must begin talking about creating a health data resource in a much broader and more universal context, controlled by the individuals who supply the data. This is a unique moment where we may be able to provide for personal control and, at the same time, create a global knowledge medical resource.

We have coined the term 'UnPatient' for our new model of data ownership as it has the double entendre of the patient subjected to medical paternalism and information asymmetries, along with the idea that it has taken far too long to become free to use our medical data as we see fit and to own it. Without connecting to their medical data, people are unnecessarily being hurt and dying. Accordingly, we urgently seek to promote ownership of one's medical data as a civil right and as a pivotal strategy to further digitize medicine, providing a new resource to potentially help every individual who willingly participates.

This is the essence of the benefits of democratization: shared control provides shared benefits at an exponential rate. When individuals inform the collective, and the collective informs the individual, we will have the learning health system we seek.
People connecting over information to create value and share assets has been the story of our economy for the past 20 years. So, too, will it be with sharing and leveraging medical data in the next 20 years.

  1. Bachman, R. The United States of Fitness. Wall Street Journal (26 May; accessed 5 July 2015). http://online.wsj.com/public/resources/documents/print/WSJ_-D001–20150526.pdf
  2. O'Brien, J. Can big data help you get a good night's sleep? Fortune (29 June 2015). http://fortune.com/2015/06/29/sleep-data/
  3. Topol, E.J. Cell 157, 241253 (2014).
  4. Xu, G.J. et al. Science 348, aaa0698 (2015).
  5. HHMI. Your viral infection history from a single drop of blood (4 June; accessed 7 July 2015). http://www.hhmi.org/news/your-viral-infection-history-single-drop-blood
  6. Cannon, W.B. The Wisdom of the Body (W.W. Norton & Company, Inc., New York, NY, USA, 1932).
  7. Mandl, D.M., Mandel, J.C. & Kohane, I.S. Cell Syst. 1, 813 (2015).
  8. The Office of the National Coordinator for Health Information Technology (ONC). Report to Congress April 2015: Report on Health Information Blocking. (Health IT, 9 April; accessed 7 July 2015). http://healthit.gov/sites/default/files/reports/info_blocking_040915.pdf
  9. Fickenscher, K.M. J. Am. Med. Inform. Assoc. 20, 593594 (2013). http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3628072/
  10. Robert Wood Johnson Foundation. Individual access to medical records: 50 state comparison. Health Information & the Law (24 September 2013; accessed 7 July 2015). http://www.healthinfolaw.org/comparative-analysis/individual-access-medical-records-50-state-comparison.
  11. Topol, E.J. The Patient Will See You Now (Basic Books; New York, NY, 2015).
  12. Kish, L. The blockbuster drug of the century: an engaged patient. HL7 Standards. (28 August 2012; accessed 6 July 2015). http://www.hl7standards.com/blog/2012/08/28/drug-of-the-century/
  13. Collins, K. A quick guide to the worst corporate hack attacks (18 March; accessed 6 July 2015). http://www.bloomberg.com/graphics/2014-data-breaches/.
  14. Beck, M. How to take charge of your medical records. Wall Street Journal. (29 June 2015). http://www.wsj.com/articles/SB12367224787933994021304581064031716335262.
  15. https://twitter.com/Farzad_MD/status/587233286230945792. Accessed 7 July 2015.
  16. Peck, M.E. The future of the web looks a lot like bitcoin. Spectrum IEEE (1 July; accessed 2 July 2015). http://spectrum.ieee.org/computing/networks/the-future-of-the-web-looks-a-lot-like-bitcoin
  17. Chu, S. Apple watch release news: survey finds 80 percent of US employees would give health data from wearables to employers. iDigitalTimes (2 February; accessed 7 July 2015). http://www.idigitaltimes.com/apple-watch-release-news-survey-finds-80-percent-us-employees-would-give-health-data-411578

Acknowledgments

Special thanks to J. Robinson, N. DiNiro and D. Maizenberg, members of the UnPatient team, for assistance; US National Institutes of Health grant NIH/NCATS 8 UL1 TR001114 for supporting E.J.T.; M. Miller for preparation; and J. Hightower for assistance with graphics.

Author information

Affiliations

1.     Leonard J. Kish is at Unpatient.org, Denver, Colorado, USA.

2.     Eric J. Topol is at The Scripps Research Institute, and Scripps Health, La Jolla, California, USA.

Competing financial interests: The authors declare no competing financial interests.

Corresponding author: Correspondence to Eric J Topol

################################################################################

VII). Also see: http://YouBase.io (A truly individual-centric data store for healthcare, owned and controlled by the individual)

YouBase White Paper

About this book:

(See: https://www.gitbook.com/download/pdf/book/joshrobinson/youbase (19 pages)

By Josh Robinson & Leonard Kish (6-1-15)

Abstract

YouBase enables individuals to create and maintain a personal data store on a distributed public network, allowing the unprecedented ability to easily gather, analyze, and share private data for any purpose imaginable. Data is structured hierarchically so that increasingly identifiable data can be placed at levels closer to the root, allowing arbitrarily anonymized data to be shared with whomever is requesting access to it. The data format is flexible, enabling easy integration with third parties. In addition, read-only or read/write access can be granted at any node in the tree, allowing the user to tightly control access to every subtree in the data store. YouBase thus provides the building blocks for the ultimate peer-to-peer central repository for private data, enabling individuals, organizations, and the world to make smarter decisions.

Introduction

Cryptography combined with distributed applications and databases in peer-to-peer networks provide the fundamental building blocks required for securing stores of individual-centered digital property in an open standards-based manner. By using encryption, digital signatures, digital wallets, and distributed data, ownership of digital information can be managed in a decentralized store. Such a store will be simultaneously secure and private, with strong identity services, while also available anywhere.
Information and rights to that information will ideally follow an individual as she moves through various contexts in her daily life, enabled with the ability to provide trusted, verified identity within those contexts. A longitudinal record could be created, including consumer-generated application data, with the individual as the primary controller of access - all independent of a third party.
YouBase provides an individual-centric security structure that separates personal data from identity while allowing for secure and structured read and/or write access to trusted parties on a peer-to-peer data store. This structure provides several benefits, including:
  1. a way to securely input, access and share any kind of file or record
  2. a way to organize authorized access to information into a structured hierarchy
  3. improved anonymous information sharing that could be used as a public or shared private asset
  4. information sharing transactions can be to be tied to financial transactions
With these tools in place, we imagine a world where, rather than storing personal data, third parties could simply subscribe to data owned and controlled by the individual.

Implementation

YouBase introduces a new type of bitcoin HD wallet which uses BIP32 public-private key pair trees and couples this wallet with persistent peer-to-peer content-addressable key-value stores, such as IPFS. The new HD wallet implementation is responsible for maintaining the BIP32 nodes defining the hierarchical structure, metadata, and pointers to the actual data, as well as managing permissions and handling encryption/decryption. We propose a distributed hash table for holding this BIP32 tree metadata as this makes it easier to access, backup, and sync. The peer-to-peer key-value store is treated as just that, a dumb key-value store, so technically it could be hosted locally, on the cloud, or on a peer-to-peer file system such as IPFS. Value would be maximized using IPFS as data would be public and decentralized.

Updated a year ago (June 2015):