# Accelerating Consumer Crypto with Application-specific Protocols

Sreeram often compares EigenLayer to a **[Verifiable Cloud:](https://x.com/sreeramkannan/status/1777791230556119183)**

When you want to build an app, you build a server.
***When you want to build a verifiable app, you build a rollup.***

The internet enables apps to communicate with each other.
***Ethereum enables rollups to verifiably communicate with each other.***

Building all the components of your app is hard, so app servers integrate SaaS / hosted solutions, which solve specialized problems.
***Building all the components of your rollup yourself is hard, so rollups can leverage AVSs to solve specialized problems in a verifiable manner.***

Hosting your server and SaaS yourself is hard, so you use a cloud provider, who leverages economies of scale to run your software.
***Running your own rollup and services verifiably is hard, so you can leverage eigenlayer to verifiably run your rollups and AVSs.***

<img
width="200"
src="https://hackmd.io/_uploads/rJMPw5cQC.jpg"
/>

### But what does this actually mean?

With EigenLayer, developers can more easily build arbitrary, and specialized, decentralized and verifiable networks.

They can also experiment more easily because the risk of building something is much lower - they are not dedicating as many resources and can also iterate much more quickly.

### Why is experimentation and specialization important?

Cost, both monetary and opportunity, often dictates what people build.

People who spend years of their lives focusing on solving a problem may choose not to swing for the fences for a 1% chance of success, and will instead go for a single in exchange for a 20% success rate.

This stifles innovation. Lowering costs enables more experimentation and therefore accelerates innovation.

#### Consumer-facing unlocks

In the context of web3, accelerated protocol experimentation unlocks application-specific features and differentiation. If you want to build something differentiating, you no longer have to rely on the constraints of the existing underlying protocols you are using, you can instead build and customize everything yourself.

A good example is Farcaster, which recently [raised $150M](https://x.com/dwr/status/1792948611585831302).

Dan Romero, the founder of Farcaster, put together a nice (short) piece titled [Product-led protocol development](https://danromero.org/product-led-protocols.html).

The idea is that if you want to build a successful protocol, you should start by developing a high-quality app that attracts enough quality daily active users to entice developers, ultimately fostering a diverse ecosystem of independent clients and applications.

But most protocols take the opposite approach, they build infrastructure and cross their fingers that a developer will come build a killer app there.

More rapid iteration at the protocol level unlocks the ability to accelerate the iteration of consumer-facing features.

#### AVS Legos

As more specialized AVSs are built, you may not even need to build that much yourself.

AWS not only offers hundreds of service, they've also enabled countless services to be built on top of them.

Over time, more and better decentralized protocols and verifiable services will become available to solve more specialized problems, allowing developers to offload the time and complexity and instead spend their time focusing on product differentiation, features, and UX.