There are 2 product building philosophies:

  1. Supernormal; check Dave Morin’s post on this
  2. If you are great, you need not be good; refer to Paul Buchheit’s post

Supernormal: If you are building a new bucket, you have to first make sure that your bucket has all the characteristics of a standard bucket. There should be a handle to carry it. It can’t have a hole that leaks water. It has to be durable. Ideally light weight. Then once you nail the basics, you make it super in at least one dimension that caters to the need of some particular TG.

  • It is a Supreme bucket: A collector item, has the logo and hence is very expensive.
  • It is made of some environmentally friendly material to cater to the people worried about their impact on the environment.
  • Maybe it is lightweight, can move easily [has wheels] and also cheap [improvements on 3 dimensions]

But it will always have what you expect from a bucket. The user does not have to change their mental model of what a bucket is. And the criticism and the praise for the new bucket design will also come from expectations on how a bucket should behave.

This is how most people design new products: They look at existing products and try to make it supernormal in one dimension.

[Note: You can build a startup by finding newer ways of distribution, changing pricing, and providing better customer support etc too. But for this post we will only focus on product development]

Then there is the other philosophy of building products: If you are building something truly great, it doesn’t need to be good.

This means that the new product will be truly great [10X better] in some existing dimensions that will compensate for the product not having the things that most users consider basic and a must-have for that product.or introduce one or more new dimension that people did not have in mind when they thought of a product.

Gmail stored all of your email, you did not have to delete your emails to get storage, but it did not have a rich text composer.

Let’s come to blockchain [take Ethereum for example]. 

I think most of the arguments between web2 and web3 happen because web3 folks feel that the idea of an

  • immutable
  • database
  • decentralised decision making 
  • no central ownership
  • transparency by default
  • financial rewards being aligned with participants from day1
  • programmable [using smart contracts]

is so revolutionary that it is fine if a CRUD call takes 15 secs on Ethereum and the call costs 100 dollars.

They feel that with time [think Eth 2.0] both will be solved. They feel that the innovations that the blockchain brings are so great [on multiple dimensions], that it does not compete with a traditional database [be good in the dimensions of speed and cost.]

[Obviously there are more arguments around privacy, environmental concerns around proof of work, the inability to reverse transactions, people getting rug-pulled in web3 projects, but we will ignore those for now and just focus on the blockchain]

While web2 folks use their super normal mental model ask how can a database be so slow and costly. These are fundamentals. They can’t move beyond underperformance in these two dimensions central to their thinking of how a database works.

Who is right? We will see.

One thing I am sure of is that we can view web3 projects using the mental model of super normal. We have to ask whether the benefits are so great that it is okay for it to be not good. And that is the hardest part.