Blockchain will create a huge impact in the future. It is a new approach in distributing information among untrusted participants where these participants need to compete and cooperate at the same time. But first, we need to define: what is blockchain?
Surprisingly, it is not easy to answer the seemingly simple question.
The query arose during a special blockchain session held during the second day of ISPEC 2017 conference. Joseph Liu (Monash University and Blockchain Joint Lab), Khal Achkar (Hcash and Blockchain Joint Lab), Tsz Hon Yuen (Huawei Singapore), Andrew Miller (Univ. Illinois and Zcash Foundation), and Man Ho “Allen” Au (Hong Kong Polytechnic Univ and Blockchain Joint Lab) were the panel members.
Well, actually it was all started by Andrew throwing the question for the session. Although I already had that question in mind, I did not have the gut to say it since I will look dumb saying such question. He mentioned that he was unable to determine what blockchain really is and it is possible to define blockchain as an inclusive term. It means one might not be able to say this thing is or is not a blockchain. We know that in Bitcoin, the blockchain looks like blocks of transaction chained one to another resembling a logical chain of data. But when we talk about possibilities of constructing different shapes of blockchain, then its meaning is really broad.
After the session, Allen approached me during the break. He said, a couple of weeks ago he attended a discussion with several people trying to standardize blockchain, hoping to formulate a standard form for blockchain as they did for cloud computing. At the end of the day, the discussion did not conclude anything.
The fact that there are various types of blockchain means that it is really hard to describe the blockchain. We commonly say that blockchain is a distributed (or decentralized) database with no central authority. But you know what, there are cryptocurrencies employing centralized consensus (or somewhat centralized) e.g. Dash or IOTA. These coins make use of “master nodes” or trusted delegations in which they need to agree upon new transactions. They might realize that total decentralisation might require extra efforts and more energy to write transactions into the blockchain.
The next point to consider is what type of database can be considered as a blockchain. I used to believe that the product used to construct the blockchain needs to satisfy certain requirement e.g. fast enough for the server to do I/O operations and therefore the standard DBMS should be unsatisfying. But then I read that LISK uses PostgreSQL as their backbone which is pretty unusual considering that Bitcoin uses LevelDB (previously it used Berkeley DB) and Monero uses LMDB. So it is inconclusive to determine blockchain based on their database product.
And for the number of servers (or nodes) hosting the system, I am unsure how many nodes are required for a system to be called as a blockchain system. What if there is only 1 server or 2 servers? It is generally believed that the security of blockchain technology lies in the number of nodes. But what if they all get hacked? It is not impossible for someone (or a group of extraordinary hackers) to hack all these. I believe Andrew mentioned that at some point in the past, someone hacked the Bitcoin network by embedding a malicious software header that if the node is equipped with an antivirus, the antivirus will block the data and it will cause chaos to the system and eventually it crashes. So, it is also inconclusive to determine blockchain based on the number of nodes.
Peer-to-peer network? Oh well, it is pretty obvious that anyone can use the P2P technology to construct a system to communicate between computers, and it does not mean that a system using P2P technology is a blockchain. Likewise, it is also inconclusive that a system not using P2P technology is not considered as a blockchain system.
So, if I need to conclude all these, I can say that I concur what Andrew and Allen said that blockchain is pretty hard to describe, in the sense that the technology is still in the early stage and not mature just yet.
What do you think? Write your comments below.