AMA with Algorand Ambassador, Harry Boer

Every week, we hold an AMA on our invite-only ecosystem slack channel, to foster conversations geared towards building trusted infrastructure for the borderless economy. If you would like to contribute to growing the Algorand ecosystem, follow the instructions at the bottom of this article and join our core team, partners, ambassadors, and top community members as we share innovative ideas, feedback, and updates on all the different facets of the Algorand project.

______________________________

AMA Transcript | October 2, 2019

Introduction

Hi all! Let’s warmly welcome Harry, our Ambassador in Amsterdam who will be sharing insights about the ‘Remote Patient Monitoring’ project, which he is developing on Algorand. Harry has been working on the Hyperledger Fabric, IOTA DAG, and has played an instrumental role acting as senior developer for the ABN-AMRO clearing bank. He is also the proud winner of the IBM sponsored Hackathon in 2016 and currently using DLT to revolutionize the health sector. Glad to have you here Harry!

We will be having 2 sessions; the initial would be a 15 min session covered by me, followed by community QA session of 15-30 minutes.

--

Harry Boer:

Hi guys, Nice to be with you and thanks for this opportunity to talk to you.

To quickly say something about myself. I’m a Geochemist by training. Later on, I did some postgrad studies in Economics and Financial management. One thing those two fields have in common is data analysis/statistics/programming. First in Python and Perl much later on in Golang. I started working on DLT related subjects after I read about Ethereum (Dapps), but very soon thereafter I developed in Golang for HLF and IOTA. When I read about Algorand I knew this was the thing to settle for as problems like scaling, speed and (de)centralization etc. etc. were solved.

Please shoot.

Session 1 

Thank you for the introduction! Could you briefly tell me what inspired you to use Blockchain for health issues and what inspired the transition to it? How has the journey been so far for you?

--

Harry Boer:

OK, just to give some background. It is a well known fact that costs in the health sector are exploding. One way of helping to reduce these costs is to release patients from the hospital as soon as possible — something most patients will appreciate as well. Yet, it will be necessary to monitor these patients in their home environment for a couple of days. This is called Remote Patient Monitoring (RPM). My partner (a medical biologist and researcher) realized that RPM will become a must for the future and that the safety of data transmitted will be crucial for future applications. On top of that the General Data Protection Regulation (EU) 2016/679 (GDPR) will impact profoundly on how patient data will have to be protected and managed. Therefore, new technologies and other ways of looking at data are needed.

RPM is an extension/subset of a much larger phenomenon, i.e. the upcoming M2M (Machine to Machine) economy. This will be driven by the expected exponential growth of the IoT. Our solution is not only for RPM, but will impact on a much wider M2M economy field. Estimates on IoT growth differ enormously, from 75 billion IoT devices to 125 billion within a couple of years. Furthermore, 80 billion devices are expected to generate 160 Zettabytes of data/year in 2025. To illustrate, 1 Zettabyte is 1 billion Terabytes.

Furthermore, according to my opinion, DLT will become a crucial tool in guaranteeing authenticity and provenance of (sensor) data.

--

What are the core beliefs that made you develop your solution on blockchain as opposed to another encrypted system?

--

Harry Boer: 

My core belief is that:

“All (sensor) data that have, or in the future could have, any value, e.g. juridical, financial/economical, scientific, medical, privacy related etc., must have guaranteed authenticity and demonstrable provenance.”

This means that we must be sure that data we’re looking at have not been tampered with (authenticity) and were derived from the expected source (demonstrable provenance).

My second belief is:

People who claim their system cannot be hacked do not fully understand the phenomenon and/or complexity of hacking. Hence, it can be assumed that their system can be hacked.

Some of them claim, “Yeah, but our data/system/whatever is encrypted.” My response to that is:  Listen Bro, if something can be encrypted, it can be decrypted.

Note that Hashing (one-way encryption) used in DLT is different as you can never retrieve information from the Hash.

Data must be immutable and decentralized to avoid SPF. (Single point of failure -- where one node going offline affects the whole system).

--

Glad to hear the explanation! It seems you are targeting cost efficiency and immutability : ) Where does the Algorand blockchain come into all of this? Would love to hear how you have utilized Algorand to achieve this.

--

Harry Boer:

The Algorand DLT is important for multiple reasons:.

First, due to the large datasets that have to be transmitted with RPM, we had to find a way to transmit the data and store them. This is not easy with ECG data. So we had to split it.

We have split the data to a PostgreSQL database. The data are hashed. (You can see the message under RQ6PBUGVNGMALPLUF6O6VIS5GJF2TS6U4QQAY3FZ6H2LPRG2F7CQ in the TESTNET).

Finally, that hash will be put on the Algorand blockchain with TxID (transaction id) which will then be linked to the data hash.

Secondly, the accounts “From To”  will be the accounts of the owner of the device (usually hospitals) and the health insurers. So Algorand will play a very important role.

Over to you.

--

Thanks! So if you were to give a very short elevator pitch on what RPM is about, what would it be for a layman user?

--

Harry Boer: 

Remote Patient Monitoring will play an important role in reducing costs in the health sector and add to patient satisfaction. As the patient can be released from the hospital much earlier than is the situation now.

By the way, RPM is not something that was invented by us, but a topic that was introduced by hospitals. We are using Algorand to design and make confidential data tamper-proof. 

--

Daniel:

Indeed, I have been hearing about RPM related topics for a while in medical circles. Are you planning to give your product a name any time soon?

--

Harry Boer: 

First of all. Our company is called Life2Ledger.

Our mission-statement:

“Life2Ledger aims to become one of the first remote, GDPR compliant, (medical) data solution providers guaranteeing authenticity and provenance of data”

--

Daniel: 

Nice choice for a name! Are you facing any competition in this area or do you consider yourself having a first-mover advantage?

--

Harry Boer: 

No, we haven’t heard yet about anyone doing the same thing as us. However, that doesn't say much as most people actively working on DLT solutions remain hidden.

Not due to any particular reason but just because they are very busy developing.

So no, I don’t know of anyone applying a similar solution.

--

Daniel:

So I actually have a question from Russell Fustino, our tech evangelist regarding the use of Algorand:

“Which Algorand SDK have you used to develop your product? How was the experience building and would you recommend any improvement feedback for us?”

--

Harry Boer: 

I know Russell Fustino, we had a beer in Amsterdam. I used the algorand-go-sdk.

He talked me into Algorand! I use the Go sdk, because that’s my favorite programming language.

It is very well documented. And the examples are very clear. I had put aside a week to put the hash and message in the note64 field, but it took me less than one hour.

Well done Jason Weathersby and team.This easiness is a great selling point.

Oh, one thing. We want to do a PoC in April having produced our first product.

The PoC will be done in a University Medical Center in The Netherlands.

--

Daniel:

I am sure Jason Weathersby would be glad to hear about the seamless experience you had with the SDK. Developing it is an ongoing process and feedback from our devs/builders helps make it better! While we are on the same thought pattern, I had another question submitted from Asaolu Elijah from Milan: He was wondering if you had any pointers for an aspiring dev who is just starting to build on Blockchain?

--

Harry Boer: 

In what language? Look, the only advice I can give is that you must have patience.

DLT is not easy. But having said that. Just try, do the examples given on the forum using the TESTNET and build form there. I must say, the Algorand team is very helpful and if you have any questions, just ask. NEVER BE SHY. Note that everyone of us had to start somewhere.

--

Russell Fustino (Tech Evangelist - Algorand): 

To add to the above question! We have been building up the devrel content on sample solutions here: https://github.com/algorand-devrel and in particular, I wrote a lab exercise for the hackathons, that can actually be used as a quick start with step by step instructions and screenshots and the code samples for out of the box with a few alternatives for stand up instances via purestake and hackathon. https://github.com/algorand-devrel/hackathon 

Community Round

Merson (Algorand Community): 

We see governments integrating blockchain tech, specifically in South Korea (healthcare) and the UAE (smart cities, DeFi). Where do you think will be the next country/region to adopt this tech?

--

Harry Boer: 

Merson that is difficult to say, but what I do know is that, for example, GDPR type of regulation will become some kind of standard.

So I’m not sure which country will adapt this technology. And oh, don’t make a mistake, This technology has NOT been adopted yet in the Netherlands, but I hope it will be soon. 

--

James (Algorand community): 

Hi Harry, what kind of regulatory hurdles do you anticipate bringing this kind of project to fruition? I like the idea of bringing blockchain to healthcare, but I know that in my country (Canada), they've been working on/talking about digitizing patient records and other things of that nature for a LONG time, but it's been extremely slow moving, for a variety of reasons (some also technical). I can only imagine how complicated it will be to implement blockchain to this industry. Any thoughts?

--

Alex Broudy (NewYork Ambassador): 

Harry it’s great to hear you’re building a GDPR compliant healthcare solution on Algorand! I’ve been keeping an eye on privacy protections rolling out in the States lately. Do you have an idea of how the CCPA (California Consumer Privacy Act), Nevada Privacy Rights or New York’s proposed privacy updates fit into the picture? Would a federal level privacy initiative make it easier to develop HIPPA (Health Insurance Portability and Accountability Act) compliant tech solutions? Thanks! 

--

Harry Boer: 

To both questions above, I have no clue about legislation in the USA or Canada. But I think that it will be similar along those lines. GDPR basically says; “Patient data belong to the patient and NOT to the hospital.” This means that a patient decides whether his/her data are allowed to be used, kept or sold.

I think the US will be embarking on a similar route. Sorry, I can’t be more specific. I’m not a lawyer.

--

Karma (Algorand community): 

Harry, Do you think your solution might be implemented not only in healthcare, but maybe in childcare, in security field and so on? If yes, which field is the most interesting for you to fork Life2Ledger?

--

Harry Boer: 

Interesting question. Childcare, obviously.

Can I give you some subjective statements of mine:

“All (sensor) data that have, or in the future could have, any value, e.g. juridical, financial/economical, scientific, medical, privacy related etc., must have guaranteed authenticity and demonstrable provenance.”

This means that we must be sure that data we’re looking at have not been tampered with (authenticity) and were derived from the expected source (demonstrable provenance). So yes, Life2Ledger can and will fork for anything that needs DLT.

--

Karma (Algorand community): 

Can it also be used in penal system? e.g. to control if people under house arrest are really inside the house

--

Harry Boer: 

This could be in the sense that DLT can prove whether a person has stayed inside or not. One example. It’s hilarious but it happened in the Netherlands. A woman working at the forensic laboratory was fined for driving with alcohol in her blood. They discovered that she changed her values in the database at the lab where she was working. That could never happen with DLT.

--

Kong (Beijing Ambassador): 

Hi Harry, I am not a programmer. I want to know what’s the difference between using IOTA and Algorand for developing your project 

--

Harry Boer: 

IOTA and Algorand are two completely different DLTs. IOTA makes use of the Tangle, which is a Directed Acyclic Graph. Algorand makes use of Linked Lists and cryptographic sortition. One thing that they have in common is that the hash of block n-1 is used and hashed together with the data of the TX in block n. The reason why my choice has ultimately fallen on Algorand has everything to do with the Blockchain trilemma.

--

Daniel: 

Which one is better between IOTA and Algorand? Did you think about using some other Blockchain first before choosing Algorand?

--

Harry Boer: 

Yes, I had a rudimentary HLF (Hyperledger Fabric) blockchain running locally with all the bells and whistles, but after my experience at the bank, I knew that this was faaaaaar too complicated for health institutions. 

--

Mike: 

Hey, Harry, are you looking to do any kind of fundraising for Life2Ledger? Any interest so far from investors?

--

Harry Boer: 

Yes, we are looking for funding. We had meetings recently with people managing a fund for the University medical center and we have a follow up meeting in two weeks time.

--

James: 

Sorry if this has been asked, but do you guys have a roadmap for this project? What kind of milestones are we looking at?

--

Harry Boer: 

Yes we have. One of the milestones will be to finish our first hardware gadget that will be the interface between the ECG and the blockchain. This will be demonstrated during a POC somewhere in April. Furthermore, we have a Gantt chart with some short term milestones.

--

Haardik (University of Waterloo): 

Can you clarify more about why HLF was not suited/what experience at the banks? HLF is permissioned private by design and Algorand is public. What trade offs did you have to make?

--

Harry Boer: 

It is too complicated. If you have to ask the client to manage an “Orderer” a “Certificate Authority” an “Endorser” with all kinds of Peers on different nodes in the network, it is asking for trouble. I can’t go into specifics. 

--

Haardik (University of Waterloo):  

What was the trade off in shifting from a permissioned ecosystem to a public one though? Layers of more encryption and more resources put into it to look for relevant transactions? Does each client run a 24/7 worker to filter out relevant events? Do they still own a private key to decrypt information intended for them?

--

Harry Boer: 

The complexity of maintaining the network. For the bank for which I was working this is not a problem, but for the fintech companies who were their clients, it was. As a consequence, all the nodes were running at the bank’s sites, thus decentralization dumped. So again, complexity was the main reason.

Yes, at first I found the permissioned solution great, also for health sector, but now...

--

Kong (Beijing Ambassador):  

I am healthy but I still use Apple Watch to help me monitor my heart beats and sleep quality. So if I’m using your product in my daily life, can I get token rewards for giving out those regular data? I believe legally data trade is also a big market. 

--

Harry Boer: 

Yes, data will become tradable and so the authenticity should be guaranteed which makes for a big market.

______________________________

That’s all from the AMA folks! We thank Harry Boer for sharing his thoughts and our community for taking part in the AMA. 

This would be part of a weekly series where you can meet our ecosystem partners and learn what they are building. If you are a builder, DeFi believer, aspiring coder or simply someone wanting to make a difference, you can join the Algorand Ecosystem slack as well and take part in many such conversations by sending your emails to [email protected] 

Alternatively, you can apply to become part of our Community through https://community.algorand.org and join our growing network of global ambassadors!