The myth of the infrastructure phase


The myth of the infrastructure phase

Within the concept of Web 3.0 is considered to be that we are in the infrastructure phase, and now is the time to work on the creation of infrastructure: improving basic circuit, interoperability between the chains, wallets and browsers. Explain this: first, we need tools that will simplify the creation and use of applications running on the blockchain, and when we have these tools, we can begin to create these apps.

But talking to people, building the infrastructure, we know that it is still the biggest problem is forcing developers to create applications. That is, if we really are in the infrastructure phase, why would that happen?

Our hypothesis is that it is actually works differently. We are not in the infrastructure phase, and at the bend of the loop “application infrastructure”. And in fact, the history of new technologies demonstrates that the applications generate the infrastructure, and not Vice versa. We don’t build all the infrastructure needed to then begin to create applications. All the way around.

Are we even talking about this mainly because now everyone knows that the “platform” are often given the biggest opportunity for earning (this is true for Facebook, Amazon/AWS, Twilio, etc.), so it is understandable desire to create a large platform that will generate income. Even more true this may be in a distributed network where the income is often, though not always, stored at the Protocol level, not the applications running on the platform.

But, as we will see, platforms evolyutsioniruet in the process of repeating the cycle “applications => infrastructure => applications => infrastructure” and rarely created from scratch.

Initially, applications are encouraged to create infrastructure. Then, this infrastructure allows to create new applications.

On the example of several major platforms we see that first there’s a breakthrough app, then this app inspires the creation of infrastructure, which simplifies the creation of similar applications, and infrastructure that enables a wide range of consumers to use these applications. It looks like this:

Applications and infrastructure develop in the flexible loops and not in the individual phases

For example, a light bulb (an application) was invented before the electric network (infrastructure). For light bulbs mains is not needed. But for General use incandescent lamps network needs, therefore, first in 1879 was light (breakthrough application), followed in 1882 appeared the mains.

Another example: airplanes (application), invented before there were airports (infrastructure). To have the plane do not need the airport. But for widespread use of aircraft the airports, so at first there was a plane in 1903, it inspired people to create the first airline in 1919, and then there were airports in 1928 and air traffic control in 1930. But at first was a plane.

The same picture with the Internet. First came the apps: messaging (1970) and email (1972), they are prompted to create an infrastructure that facilitates the wide dissemination of messaging and email; Ethernet (1973), TCP / IP (1973) and providers of Internet services (1974). Then came the next wave of applications — web portals (Prodigy, in 1990, AOL in 1991), and web portals inspired the creation of the infrastructure (search engines and web browsers in the early 1990-ies). The next wave of applications, websites such as (1994), they have pushed for the creation of infrastructure programming languages (PHP in 1994, JavaScript and Java in 1995), which simplified the creation of websites. This was followed by a new wave of more sophisticated applications, such as Napster (1999), Pandora (2000), Gmail (2004) and Facebook (2004), which led to the creation of an infrastructure that facilitates the development of more complex applications (NGINX and Ruby on Rails in 2004, the AWS in 2006). And the cycle continues.

This is the same pattern we see in the field of mobile applications: for example, the first was a collection of popular mobile applications, which relied heavily on streaming video: Snapchat (2011), Periscope (2014), Meerkat (2015) and Instagram Stories (2016). And then, the company has created an infrastructure that allows mobile applications to add video: Ziggeo (2014), (2014), Mux (2017), Twilio API Video (2017), Cloudflare Stream (2018).

This cycle allows you to properly arrange and order events in Web 3.0. We started with the first breakthrough application — BTC (2008) working in the Bitcoin network (infrastructure). Soon came the Silk Road (2011) — the infamous cryptographic application. This prompted to create a new infrastructure, such as Sidechains and Drivechain (2015), an Ethereum Smart Contracts and ERC20 (2015), Lightning (2015), allowing you to easily create new applications and infrastructure such as Coinbase (2012) and Metamask (2016), allowing consumers to use these new applications. The new infrastructure has spawned the next wave of applications: tokens/ICO (2017) and the first decentralized applications (Rouleth and vDice in 2016, CryptoKitties in 2017), which led to the creation of new infrastructure: Infura (2016) and Web3js and Zeppelin (2017). It now remains to wait for the next visible application, which will serve as benchmarks for new infrastructure wave.

Related opportunities

Shared in development of every major platform (be it electricity, automobiles, airplanes, the Internet or mobile devices) that we build what we can, using the tools that are available to us at this moment. In the book Where Do Good Ideas Come From (“Where good ideas come from”), Steven Johnson calls this the “adjacent possible.” In other words, you can open the door to the next room, but you can’t just open the back door from the front porch. It is difficult to build a successful infrastructure, if it is much faster than the app market.

At each iteration of the loop “applications => infrastructure” creating new applications become possible thanks to the infrastructure built in previous cycles. For example, YouTube could occur in 2005 but not in 1995, because YouTube makes sense only after the deployment of infrastructure, such as broadband in the early 2000-ies. It happened in the infrastructure phase that followed the rise of successful “dot com” sites, such as eBay, Amazon, AskJeeves and my beloved Neopets.

Chris Dixon and Fred Wilson talk about this concept in a recent episode of the a16z podcast. Chris is a Board game of days of the dot-com called Dot Bomb, which mocked the ridiculous dot-com of the late 1990s. So here it is noticed that all the “absurd” ideas of the era of the dot-com today have become “unicorns” with billions in turnover. What is possible now, after a few cycles of “applications => infrastructure” in the Internet, had no sense of one-two cycles ago.

This is the main problem of the myth about the infrastructure phase: if you think about the “infrastructure phase” in isolation from the applications that will use the infrastructure, we risk to run too far, in a hypothetical void. For reliability we need cycle “applications => infrastructure => applications => infrastructure”.

The more cycles each new platform, the cheaper to create and use these applications. Create in 1995 would have cost us orders of magnitude more than today, and the creation of Web 3.0 requires more money, effort and time now than in 15 years.

Technological principles vs investment

Try on the role of investor. It is important to distinguish technological principles that explain when something can be created, and investment that determine when something can make good money.
Cycle “applications => infrastructure => applications => infrastructure” explains when you can be created by applications or infrastructure, but not necessarily explains when to invest in applications, and when to invest in infrastructure.

Take, for example, light bulbs. Yes, they were invented before the emergence of the network, but in the context of investment, disgusting bulb sold until they built the grid.

Summing up

Main question: why cycle begins with the applications, not the infrastructure? One of the reasons is that it makes no sense to create an infrastructure before there will be applications with the need to solve their problems with infrastructure. As you can see that the created infrastructure will solve a real problem, yet applications for which these solutions are used? It is hard to create Cryptoprotected, yet the breakthrough cryptopidonia that other developers will want to imitate what they need more effective development tools and infrastructure.

In Cryptoprotected of the opinion that first you need to create great instruments and when they are, we will be able to create applications. But, as we demonstrated with examples from other areas, it is necessary to create the first few applications before it will appear great tools (although this is more expensive and for the money, and time), and then these early applications will inspire the creation of tools. And then the cycle will repeat.

Good luck in development!


Update (October 5, 2018).

Was correctly noted that kryptonite, in fact, blur the boundaries between applications and infrastructure due to its nature of openness and interchangeability. And that’s what we like best. For example, “application” (whether CryptoKitties or any smart contract or the bitcoin) can be the infrastructure, if someone develops based on it. Of course, there are components of these networks that is only infrastructure (Lightning, Zeppelin, etc.), but the border is blurred. At that time, as earlier platforms (i.e. Amazon or Facebook) had to make a conscious decision to open APIs and a platform, cryptographic applications usually open and compatible from day one. It only shortens the cycle “applications => infrastructure => applications => infrastructure”. We thank Denis Nazarov and uttu Steiner for the fact that he covered this point.


Original text in English here.

Leave A Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.