Your browser is ancient!
Upgrade to a different browser to experience this site.

Skip to main content

Fintech

Introduction to Blockchain

Andrew Wu gives a quick overview of blockchain, how they use data, and how they might be helpful.

Excerpt From

Transcript

This video has one goal, explain at the 30,000-foot level, what is a blockchain, and why they might be helpful. I'll share my view on this, which is not overly

technical and which I find useful in highlighting

its key features. If you're like me,

you probably first heard of the blockchain

technology through Bitcoin. Then you might have

learned something more about Ethereum, Ripple, and the large number of other cryptocurrencies

currently trading on the various exchanges. All of them essentially use

the same blockchain model, implemented using

different parameters. A blockchain is for

all intents and purposes, a

de-centralized database. As a user, you can

upload some data to it, and just like a database, they'll maintain the data and share it with whoever

you authorize, from the entire Internet

to a single sub address. This database has

two unique features that set itself apart from, say, a typical centralized

SQL based system like a web file server. First, once the data is uploaded, it's stored and arranged in a rather unique way, and second, because the system is

often decentralized, some type of consensus

algorithms have to be employed to make sure

that at the end of the day, everyone in the system received the same copy of the data

and there's no conflict. There could be a coin or a token floating on the blockchain, like Bitcoin to facilitate

value transfers, but that's not always necessary, and the token is just the

type of data that you can upload to the

blockchain "database." We'll go over all these

features in subsequent videos, and you'll see that the key

distinguishing feature of different blockchain

systems like Bitcoin versus Ethereum is number one, the type of the data that you can "upload" to the blockchain, and number two, how consensus is reached between parties

in the network. The stated goal of most

blockchain innovations is usually to achieve

some level of trust and a system integrity

in some type of financial or physical

transactions without involving a central

intermediary like a bank. Instead, the task of

the bank will be spread out to the network of nodes

comprising the blockchain. Usually, in the

marketing pitches, you'll hear several

key buzzwords, like fast transaction

verification, and ownership tracking,

data immutability, tamper resistance, and so on. As you'll see later

in this module, some of these claims

are true, some are not. So let's have a

high-level illustration of how a blockchain

works graphically. I'm sure that most

of you have used a cloud-based document

storage system like Dropbox or Google Drive, which are essentially

centralized databases maintained by these companies. Suppose you write

a Word document in Google Doc that you want to share it with some colleagues. The document is just a piece of text data that you generate

using a client application, say a word processor on

your computer or a phone. Once you write it and upload it, it goes into say,

Google's file server, which receives it

and processes it by converting it into a file format that's compatible

with the database, and once the processing is done, the data is then stored

in the database, using some particular format, say SQL or Hadoop, which use particular

file types designed to facilitate fast searching

and fast data retrieval. A blockchain does a

very similar thing using a different process. Let's go back to the beginning. So you want to upload some data on the blockchain instead. As usual, you use your client application

to generate the data, and just like your

word processor, the client app generating

these data like a Bitcoin wallet is usually

not part of the blockchain, and the next step is where

the key differences begin. Instead of uploading it to Google server, on a blockchain, you're essentially

broadcasting your data to a entire peer-to-peer network

of connected computers, which are called nodes,

and collectively, these nodes takes on the role of the single

database server. They first receive the

data in a peculiar way, then collectively

"process" the data, using the consensus algorithm,

and once processed, the data will again be

stored on the blockchain "database" using the unique

chained data structure that we mentioned earlier. We'll you use four

videos to look at each of these four

steps in detail, in the order of

conceptual difficulty. We'll first look at

the client side, that is the client identities

on the blockchain. Then, we'll talk about the

unique chain data structure, which again makes data

searching easier. Third, we'll look

at how the network receives the data, and finally, we'll look at how the

network processes the data, using consensus, which is

often the most difficult part.