In Decentralisation: the next big step for the world wide web, the author Zoë Corbyn explained why. So I will focus on "how".
Whether you open a web browser or a mobile app, what you see running in front of you is just a client. The majority of the computational workload happens on remote servers. Developers write code, then upload it to some rented servers inside some giant data center.
This is a typical request-response model. Even if you're chatting with a friend over an instant messaging app, the traffic is likely routed through a server somewhere in between. I wrote a blog post explain the 101 of cloud computing here. It's a good primer if you're very new to Internet technologies.
In the current cloud computing model, trust is needed between developers, cloud service providers and the end client. Larger Internet service providers are usually considered more trustable than smaller companies. Especially when you're going to share some sensitive data, you usually double-check to see if the website you are using is the actual site and not an impostor. There are many phishing sites out there collecting user data illegally. But how do you know those brand names will not do evil? Well, speaking frankly, they do and always have. The Internet business model relies on monetizing user data. That's how the Internet giants make so much money by providing free services to their users. We'll discuss the new business model in the decentralized era in a separate article. Let's focus on trust for now.
Let's assume a common usage case:
This is a typical problem the TEA Project is trying to solve. TEA's goal is to build a platform that allows code and data to runs inside trusted TEA modules without needing to trust anything else. The technologies built into the TEA platform protects the data and makes sure:
TEA made this possible through the following factors:
When a client deploys data to the TEA network to run on code provided by developers, nobody knows which TEA module will run the task in the future. The result will soon appear in the blockchain along with a series of PoT (Proof of Trust) data. Due to the token economy incentive, some others TEA nodes may have done the verification for you before the result appears on the blockchain. If you trust the blockchain, you can trust the result. You can also trust that no information was breached during the workflow because a zero-knowledge algorithm was applied to the task.