Scientific advice about the blockchain

The British Government’s Chief scientific adviser, Sir Mark Walport, has published his Government Office for Science report on “Distributed Ledger Technology: beyond blockchain”. In his report, Sir Mark focuses on a particular kind of distributed ledger, the bitcoin blockchain, and attempts to explain it to the general reader and then explore some of the potential uses. I’m particularly interested in his ideas about where it might be used in government, so I took the time to read through the report to examine, and learn from, his exploration.

A new report from the UK Government Office for Science has recommended a broad government effort to explore and test blockchain and distributed ledger technology.

[From Report Urges UK Government to Test Blockchain Tech – CoinDesk]

Personally, I found the report slightly confusing because it was jumping between ledgers, blockchains, the bitcoin blockchain and bitcoin almost on a paragraph by paragraph basis. I realise that I read the document from a very technical perspective and that I may see some of these things therefore in the wrong context, but I prefer Richard Brown’s term “shared ledger technology” as a starting point because I feel that the fact that multiple organisations share the ledger is more important than its architecture. I think the report might have benefited from some more description of shared ledgers, and the reasons why Moore’s Law and falling communications costs have made the core idea of everyone storing every transaction a plausible architecture. Here’s the way that we think about these things.

Birch-Pannifer-Parulava Four Layers

To be completely honest, I think our way of thinking about shared ledger technologies works well for a general audience. We use this layered approach to explain the key components of a shared ledger and then develop ideas around different choices in those layers. Different choices in consensus technology, for example, lead to a variety of different possibilities for implementing a shared ledger. In order to help categorise these possibilities, and narrow them down to make useful discussions between the strategists and technologists, we use a taxonomy that distinguishes between public and private ledgers. Rather flatteringly, Sir Mark uses a simplified version of the Consult Hyperion model of ledger technology (on page 19). However I think the report simplification is misleading in its classification and I’ve already had a couple of comments about this so I thought it would be useful to present the original model that we put forward one of the workshops that were input to the report. Here it is.

Birch-Brown-Parulava Figure 1

It might be considered reckless to disagree with the Chief Scientific Adviser, but on one matter I certainly do as the report refers a number of times to the use of bitcoin in a payments context but I just do not see this and, as I’ve written before, I certainly do not see cryptocurrency as a sensible government option for digital currency. Sir Mark says that permissioned ledgers (i.e., not the bitcoin blockchain) are appealing for government applications and I’m sure he’s right about this, although I am sceptical about some of the suggested government uses that are based on costs or efficiency. I think that his suggestions around applications that focus on transparency are the more interesting areas to explore in the short term and they would be my focus if I were looking to start exploratory or pilot projects in the field. I share the Open Data Institute’s view on this:

We agree that blockchains could be used to build confidence in government services, through public auditability, and could also be used for widely distributed data collection and publishing, such as supply chain information.

[From Comment: Blockchain technology is useful, but not for everything | News | Open Data Institute]

Anyway putting my nerdy criticisms to one side, Sir Mark’s conclusions (which are essentially that the technology is worth exploring in government contexts) are surely correct: shared ledger technology is a genuinely new way of doing things, and it will certainly lead to new solutions for the government just as much as for business. As for the government applications, there report focuses on five areas: protecting critical infrastructure, welfare, international aid, innovation and VAT. As I mentioned before, I am not sure that the areas that relate to payments should be the immediate focus. I can see that using replicated distributed shared ledgers would add robustness to critical infrastructure and there are certainly applications where a robust and immutable public record is highly desirable. I was fortunate to be asked to chair techUK’s “Blockchain 101” session last week, for example, where John Sheridan, the Digital Director at the National Archives explored just this point.

John Sheridan

Payments stuff is less clear to me. When it comes to VAT, for example, the report also mentions machine learning and quantum computing as techniques to reduce VAT fraud. I am in no position to judge but I would have thoughts that getting rid of cash would be a decent first step in closing the tax gap while we wait for quantum computing to track down the builders’ baksheesh. I do agree with the point about transparency though because sharing ledgers with regulators (and tax collectors) could help in a number of different ways.

When it comes to welfare I think that’s some of the ideas mentioned relating to digital identity are well worth exploring. There is a clear relationship between social and financial inclusion and this pivots on having a better identity infrastructure in place. It is entirely plausible that some form of bottom-up digital identity infrastructure that is built on shared ledgers could deliver better results than the identity infrastructure that we have at the moment or another attempted a top-down national identity system (while I am wholly in favour of a national entitlement system, that’s another topic).

When it comes to international aid, I suspect that once again the transparency aspect is of more practical use than the payments possibilities although obviously the high cost of international payments is an area where shared ledger technology could make a difference even if bitcoin does not. The idea of some form of cryptocurrency as a financial inclusion mechanism I think is something of a red herring. For the majority of the world’s population, M-PESA and TigoPay, AliPay and Zaad are much better ways to bring financial services to the excluded. Shared ledgers might have more of a role to play in the background, bringing interoperability into this space, something my colleagues have explored in this past.

But on to the last and most important point. Where I do strongly agree with the report is when it talks about market friction and innovation. Sir Mark highlights the potential around smart contracts and asset registration, and in our work for a number of clients (not only in financial services) this is the area that we have identified as most likely to be subject to longer term and more disruptive innovation. Remember the “4Cs” layer model at the beginning of the post? That top layer, the “contracts” layer, is where the unknown unknown lay. The work being done by companies such as Ethereum and Eris is only just beginning to explore the new domain of distributed, trusted applications and I don’t see why the government won’t be able to take as much advantage of this as anyone else.

So. In conclusion. Take the time to read the report (which kindly references my recent book “Identity is the New Money”, but that’s not why I think you should read it). There’s a lot of good stuff in there and it touches on many areas where the thinking is only just beginning. It’s great to see the technology being explored and taken seriously. Oh, and by the way, I wrote a piece called “it’s time to take stand against all the blockchain crap out there” for Finextra today, so read that too!

Wombling free

You all watch Only Connect, right? I was amazed to see a question I knew the answer to due to our nerdiness in specifying the software for Transport for London’s Tri-reader that works with Oyster, contactless EMV and even ITSO.

The question was, what do the following have in common (see image)?

Only Connect TfL Underground

Four clues are revealed one by one. The sooner you supply the link, the more points you get. All four were revealed and no-one knew the answer.  As the image shows, London Underground gate error codes, was the answer. COME ON GUYS!

Technology roadmapping

In 2005 when we performed an update to our biometrics and identification technology roadmap for the UK police, body odour was a ‘technology’ that was looking interesting, but not mature enough. The idea was that if dogs can do it, why might it not be automated. And identical twins have a unique smell, apparently.

Police biometrics techs 2005

We identified policing applications of biometrics and identification technologies, one of which was automated identification of police officers. At that time, each Force had it’s own warrant cards (so there was no confidence in what they should look like) and there was no way of using them with machines to authenticate the cardholder as an ‘officer of the lieu’ and grant them access to building and machines.

Automated identification of police officers

We foresaw the benefits of a national police warrant smart card and were retained to specify the standard which is used today across the Forces.

More recently, the technology roadmapping I have been involved in has been for transport applications. As well as the usual technologies in this space (mobile apps with 2-D bar-code; contactless payment cards; NFC mobile devices emulating contactless cards) we have also been thinking about more interesting stuff. Such as USB contactless readers used at home for fulfilment of tickets or value direct to smart cards. Or mobile devices with Bluetooth Low Energy (BLE) interacting with beacons waking the app up to present the appropriate form of ticket for the time and place. And, or course, NFC devices with the Host Card Emulation (HCE) API allowing them to escape the tyranny of the Secure Element (SE) and Trusted Service Managers (TSMs).

You’ll not be surprised to hear that we are still tracking the technology of person identification via body odour. I look forward to being sniffed by a transit gate before being allowed onto the train platform in the near future.

Managing the join

Since 2008 we have been working with Transport for London to allow contactless payment cards (CPCs) to be accepted wherever Oyster cards are accepted. This was first achieved in December 2012 on buses (which are flat fare in London) using a retail payment model. The next step was to introduce a distance-based payment model to allow all the other transport modes to be included which have zoned fares. This was launched in September 2014.

All the convenience of Oyster (such as not having to queue to buy tickets and fares capping so that you do not need to understand the fares structure) but using a card already in your pocket. Whether you are local or just visiting. But this is for London only. And the solution is based on a risk model that knows the maximum charge for a single journey is not very much. The delivery of such a solution relies on the intelligence migrating from the card to the back office. TfL’s back office to allow acceptance of CPCs for transit is complex and took several years to build.

In early 2012 the TfL payment and security models for contactless payment card acceptance in London where pretty much complete and the rest was ‘mere implementation’. TfL asked us to help them consider how it might work if they offered their back office as a service to transport operators outside of London. These might be in the UK, or potentially anywhere in the world (though different payment model are likely to apply outside of the UK). We discussed at length the notion of using your ‘card as a token’, be it a payment card, Oyster, ITSO or, potentially, other secure contactless tokens. Eventually, the ideas were parked to allow TfL to focus on delivery of the system for London in conditions of extreme austerity.

Meanwhile, we were hired by the SEFT (South East Flexible Ticketing) programme to specify the rail validators that could accept ITSO as well as contactless payment cards. At the time, Transport for Greater Manchester was just starting to procure such a back office for their region. We pointed out to SEFT that this CPC back-office-for-tranist stuff is complex and not standardised. It was therefore decided to not include any interfaces to the payment card back office at that time and the SEFT validator specification was ‘mothballed’ for the time being.

Spare a thought for the traveller buying long-distance rail tickets that include travel within the London area. London supports Oyster and CPCs (and a few specific train operator ITSO products, but not many at this point in time). Some train operating companies are implementing 2-D barcode, and some are trying ITSO. But the only technology commonly read across the UK currently and for the foreseeable future is the cardboard ticket with magnetic stripe. Basically, any ticketing innovation is scuppered at the boundary between London and the rest of the UK. This problem is what our friends at Trainline call ‘managing the join’.

Hopes for contactless payment being accepted for transit outside of London were recently dashed with the announcement that Transport for Greater Manchester has sacked their back office supplier. And anyway, it has been speculated that CPCs only work within London because London is a special case and it could not work anywhere else because the operators will not co-operate and/or the fares are too high for the risk model to work.

Enter the cavalry in the form of the UK Cards Association. They are leading a project with the Department for Transport and others (including representing train and bus operators) to develop a contactless transit framework for the UK by the end of 2015. The project to date has identified three contactless transit models:

  • Standard retail model for transit: pay as you go model with a known fare, for buses and trams (like TfL bus retail model).
  • Contactless for transit model: pay as you go model where the fare is aggregated at the end of the day or journey leg, for multi-mode operators (like TfL distance-based model).
  • Card as Authority to Travel (CAATT) model: pre-purchase model.

This last model could be just what we need for ‘card as a token’ or ‘managing the join’ as we have called it. The idea is the customer:

  1. Purchases their ticket online and associates it with their CPC.
  2. Can view their purchase on their statement.
  3. Uses their CPC as their ticket on a train.

Watch this space …

Is it safe? Is… it… safe?

Someone I know to be an impeccable source and a first-hand witness told me an interesting story about a young female friend who fell ill on holiday in North Africa. When she got home, she claimed for the doctor’s bill on her travel insurance. The claim was rejected because the person who treated her wasn’t actually a doctor. It was, as it transpired, just some guy who worked for the hotel (and presumably examined young women as a hobby). Which led me to think: how would you know? If I got sick in New York and asked the hotel to call a doctor, I’d be reasonably confident that the concierge would call an actual doctor rather than a friend who drives a taxi but has a stethoscope in the trunk. But would I check? Would I have called the New York state medical licensing board (or whatever – I just made this up) to find out?

Maybe a smartphone app that lets you take a picture of the “doctor” and then, after a few seconds, shows you a picture of his diploma would do it. Which reminds me of the old Robert Schimmel joke about going to the dentist: “Do you want a shot of novocaine / No, I want a shot of you getting a diploma”. But, for reasons related to discussions earlier in the week, I’m not sure about “passive” credential services like this. Perhaps a better solution would be that the doctor arrives with a smart card or his or her mobile phone or a badge or something else with NFC or a contactless interfaces, you read it with your phone and your phone displays a blank screen if the person isn’t a doctor and a their picture if they are a doctor with a valid license to practice in the location where you are scanning.

A woman has been charged with fraud after allegedly pretending to be a doctor at GP practices across the country…. The 29-year-old, from Maidstone, in Kent, allegedly had no medical qualifications but was thought to have used a name and registration number with the General Medical Council belonging to a real doctor.

[From ‘Fake locum GP’ who worked in practices across Britain charged with fraud | Daily Mail Online]

Now, if I were a medical practice employing a doctor, I might be tempted to at least look them up on LinkedIn or something before I let them get their hands on a patient but I suppose that under the National Health Service it’s considered ungentlemanly or discriminatory or just plain rude to ask a prospective clinical employee for verifiable evidence of any valid qualifications. We are English, so we take people at their word. Dictum meum pactum.

But then I was thinking that if I go to see a doctor for some antibiotics I don’t care if it’s a real doctor or not so long as they can write me up some amoxicillin. Or if I am expecting intimate examination for my problem, I might not care who Doctor X actually is, but I do care that they are a doctor. That’s a different problem. Anyway, being English, I am far more terrified by fake dentists.

A bogus dentist who earned almost £230,000 by using a fake degree certificate to land work at a string of NHS hospitals was jailed for three years today.

[From Bogus dentist who earned £230,000 operating in NHS hospitals jailed for three years | Daily Mail Online]

Remember, these news stories (and believe me, they are far from unusual in this sceptr’d isle) are telling us about the bogus doctors, dentists, nurses and surgeons who got caught. There simply must be others working here, today, undetected. Aargh!

Since no post on fintech right now is complete without a blockchain reference, here’s a straw man for comment. Hospitals, clinics, GP surgeries and pharmacies around the country are chock full of PCs that are doing nothing for most the time. Make them mine a blockchain of medical professionals that anyone can look up. Then when you graduate medical school you could be given a smart contract that contains your license to practice subject to certain conditions that the contract can check for itself. When I go to see the dentist, I can ask him to whip out his smartphone and demonstrate ownership of the private key that the smart license has been sent to.

Voting and common sense

Well, it’s General Election day today in the UK so I’ll be off down to the local polling station to cast my vote later on. This may be the action of a dying breed. Something like a sixth of the votes cast at this election will be postal votes and there are calls to allow people to vote at home using their PCs or smartphones just like they do for “Britain’s Got The X-Factor On Ice” or whatever it is that the general public watch on their televisions now. I’m not a fan of this kind of electronic voting.

A hacked election, or worse still, an election in which online voters buy and sell votes, would be a disaster for democratic legitimacy.

[From On internet voting, rescuing migrants, Malaysia, Turkey and gas, Scotland, the Miliband brothers: Letters to the editor | The Economist]

Well, a hacked election would be a problem, or a least an election that is hacked more than it is through postal votes right now. But buying and selling? I’m not so sure. I don’t see the ethical difference between “vote for us and we’ll give you free childcare” or “vote for us and we’ll ring-fence your pension” or “vote for us and we’ll push up your house price” and “vote for us and we’ll give you £10”. But that’s not my point. My point is that equating electronic voting with a lazy alternative to the polling booth is the wrong way to look it at.

You argue that allowing online voting in Britain would increase the number of youngsters who participate in elections (“Apathetic fallacy”, April 18th). But where is the proof? Actually, the evidence is that internet voting does not increase voter participation.

[From On internet voting, rescuing migrants, Malaysia, Turkey and gas, Scotland, the Miliband brothers: Letters to the editor | The Economist]

Frankly, if someone can’t be bothered to get off the couch and go round the corner to vote, I’m not sure I should care what they think about the way that the country is going to be governed for the next five years. That’s not what electronic voting should be about. Electronic voting should be about process re-design, modernisation and re-implementing democracy for the post-industrial age.

Is there a way to use technology to improve democracy — not only by changing the medium but by rethinking the whole interface? Well, there might be. And it is a brand new idea — in humans, at any rate.

[From Humans are doing democracy wrong. Bees are doing it right » The Spectator]

This is where electronic voting can help. Not to deliver voting by text message or WhatsApp but to deliver a new and better voting system. Now, I’m not qualified to say what that system should be, although I can see that there are many interesting alternatives to our “one man, one vote (if he can be arsed)” system.

Under Quadratic Voting (QV), by contrast, individuals have a vote budget that they can spread around different issues that matter to them in proportion to the value those issues hold for them.

[From Humans are doing democracy wrong. Bees are doing it right » The Spectator]

In an industrial age, a simple cross on a ballot paper made sense. Today, however, it should be possible to implement more sophisticated democratic systems to deliver a more accurate mandate but with simple user interfaces for everyone to use. QV on paper while standing in a polling booth might be daunting, but if you can download the smartphone app and then spend as long as you like messing around allocating your vote budget before taking the phone down to the polling booth to deliver your secure and cryptographically-protected votes via Bluetooth Low Energy (BLE) from your Trusted Execution Environment (TEE) is a different matter. Just as I always thought about eCash in the only days, Bitcoin might be more useful for voting than for paying.

Well, technology can make it easier to vote. But if there’s an app for that, we should still make people vote in public.

[From Yes, we should make voting social, mobile and local]

And to finish, one of those songs you saw on “Top of the Pops” as a kid that blew you away. I still have this album on my iPhone and I still listen to it around once every month. The best song about democracy EVAH.

Happy election day everyone.

You’ve been fingered

[John Elliott] A project we worked on for the Police IT Organisation last year is just going live in some UK police forces http://news.bbc.co.uk/1/hi/technology/6170070.stm. We worked on the business case and when we built the cost-benefit model, it was one of the most dramatic examples of a “no brainer” that I have ever seen. Fingerprint suspects on encounter and determine whether they are known criminals in 15 minutes, or take them down to the station and risk wasting, on average, four hours of police officer
time if the encounter results in release of the suspect.


Subscribe to our newsletter

You have successfully subscribed to the newsletter

There was an error while trying to send your request. Please try again.

By accepting the Terms, you consent to Consult Hyperion communicating with you regarding our events, reports and services through our regular newsletter. You can unsubscribe anytime through our newsletters or by emailing us.