What does ‘scalable’ mean, and why is it important?
Let’s say you want to run an infrastructure on blockchain. A medical information exchange system, for example. A blockchain-enabled tollbooth system for cars. Or a certification system for spare parts in aircraft maintenance. If you build such an infrastructure, of course you will want to have many people, organizations and things on your infrastructure. This means that your infrastructure will have to be able to handle many transactions per second (TPS). This is what I mean by “scalable”: “being able to handle many TPS”. And it is important because otherwise your blockchain-based infrastructure will only work for a handful of people, organizations, or things.
Why is ‘scalable blockchain’ difficult?
The problem is that you can’t have it all; a blockchain that is (1) secure, (2) decentralized, and (3) scalable. Both “secure” and “decentralized” pull in opposite directions of “scalable” because both require effort that slows things down.
This phenomenon is called “scalability trilemma”, by the way, a term coined by Vitalik Buterin, co-founder of Ethereum. There is a really good introduction to blockchain scalability by Torsten Szabolcs Sándor, here:
Currently, blockchain-based infrastructures typically are way behind other, more traditional systems with respect to scalability. Unless they sacrifice “secure” and/or “decentralized”.
First, some general observations
I wanted to know more about who is working on blockchain scalability. Companies, research, experts, etc.. So I used Mergeflow to find out.
Before I started on blockchain scalability, I wanted to get an idea of the overall momentum of blockchain in general. There seems to be a “the hype is over” feeling in some places, for example here or here. When I looked at the share of blockchain-related contents in our “industry news” and “technology blogs” data sets, it did seem to confirm the “hype is over” feeling:
But when I looked at our “scientific publications” data set, things looked different:
So perhaps what we see is more of a “thank goodness, the hype is over, now let’s get to work and build some serious things” mood. Anyways, I think that these data show how important it is to look at more than one signal if you want to form an opinion of where a technology field is moving.
Blockchain scalability: A 360° view across R&D and business
OK, time to be more specific. First, I zoomed in to our “investor-backed companies” data set, i.e. information on venture deals or ICOs discovered by Mergeflow in news, press releases, and so on.
Venture- and ICO-backed companies
There are a number of venture-backed companies that explicitly focus on blockchain scalability (without sacrificing “secure” or “distributed”). Here are three examples:
ThunderCore just raised $50M in venture funding this year. At the core of their technology is a new protocol developed by Rafael Pass and Elaine Shi, who both work at ThunderCore, and teach at Cornell University. Here is a whitepaper describing their protocol:
In their latest funding round, Algorand raised $62M. They use a proof-of-stake blockchain (which is more scalable than proof-of-work), and they want to explicitly address the above-mentioned scalability trilemma head-on. They were founded by Turing Award winner Silvio Micali (his fellow Turing Award winner Shafi Goldwasser is an advisor at Algorand).
Fusion has raised $50M in an ICO. Their focus is on financial transactions (which obviously require high TPS, if you think large scale). In line with their focus on finance, they are located in cities like New York, Singapore, and Shanghai. And their advisory board includes (former) executives from Goldman Sachs, Lehman, Qatar Bank, and others.
Now let’s turn to a different data set, technology blogs.
Tech blogs: healthcare on blockchain
Above we saw that the topic of blockchain overall seems to be declining in technology blogs. Quite markedly actually, since the first half of 2018. Here is the chart again, only with tech blogs this time:
Things are almost back down to “pre-hype” levels.
Interestingly though, when I zoomed in on tech blog posts about blockchain scalability, almost half of those were about blockchain applications in healthcare. And most of these blockchain-scalability-healthcare blog posts came out at or after the peak of blockchain overall, which you can see in the chart above at around 04-2018.
Probably not very surprisingly, these blog posts mostly talk about how blockchain could improve the handling of medical data, such as patient records. There are posts such as this about Digipharm, a company that makes a blockchain-based smart contracts platform for healthcare providers. Or this one here about Inkrypt and Translo, two startups spun out of Harvard Innovation Labs that build blockchain-based platforms for decentralized, patient-controlled sharing of biomedical data.
As we saw above, the momentum of blockchain generally in science publications is rising. When I zoomed in on blockchain scalability, “throughput” was one of the concepts or keywords that stood out.
I used Mergeflow’s graph tool to analyze the social network of researchers around this concept. Here is an excerpt of the network:
This researchers social network shows co-authors, rather than who cites whom.
If now we really wanted to get into the weeds, we could further zoom in to see the research. If we did this, here are some of the things we’d find:
Aleksandar Kuzmanovic: Net neutrality: Unexpected solution to blockchain scaling.
Changting Lin et al.: Rapido: A Layer2 payment system for decentralized currencies.
Kazım Rıfat Özyılmaz et al.: Split-Scale: Scaling bitcoin by partitioning the UTXO space.
Let’s stay in the realm of R&D but switch from publications to publicly funded research projects now.
Publicly funded research projects
One Mergeflow data set collects information on publicly funded research projects. This includes projects in the EU (currently usually under the Horizon 2020 program), Innovate UK, a research funding program by the German Federal Ministry of Education and Research (the website of this program is so bad that I’m too embarrassed to share it here), NSF, and SBIR.
Here I focused on SBIR (Small Business Innovation Research, a US federal funding program). SBIR funds individual companies rather than the big consortia that the EU typically funds. The advantage of individual company funding is that there is one point of contact for potentially interesting research (rather than first having to go through 20 or so project consortium partners before I can find out who actually did what in the project).
There were three SBIR-funded companies explicitly focusing on making blockchain scalable. Two of these companies are in the energy sector:
What’s my lessons-learned?
The first thing I learned (again; seen this in other topics before) was that “is the hype over?” kinds of questions require a 360° view. Only then can you really see how a topic develops. You can’t see this from looking at just one dimension, even though this still often happens. For example, very often “most innovative” simply means “most patents” or “most research papers”.
Then I learned that there are some very serious people working on blockchain scalability (Turing Award winners, for example). So, next time somebody says something like, “blockchain will never work because it will never scale”, and I just know that somebody will say this again some time soon, I can present them with some evidence and respond, “perhaps let’s wait and see.”