gm creators!
I hope all of you are having a great week 😎
Today's question:
How does the increasing speed of software development change the way we view competition for the next generation of client wars?
If you're new to The Bigger Picture, welcome! In this newsletter, I share insights from tech history to help paint a clear picture of what's happening in crypto today.
Subscribe below so you don't miss any future TBP posts 🥂
Do you also use Chrome?
Over the last few weeks, I've been noticing many different Farcaster clients pop up on my timeline. If you're unfamiliar with the term, a client is essentially a software program that lets you interact with data stored on a server(s). In this context, different clients provide unique ways for you to engage with the content stored on the Farcaster protocol.
I've really enjoyed using Herocast and Fiids. Fiids provides a vertical panel layout (similar to Tweet Deck) to browse through different channels concurrently. And Herocast presents your feed in a more task efficient way - how can you quickly browse, comment, and post? Think Superhuman but to get through your notifications and news feed!
This got me thinking...what will the future of Farcaster clients look like when there are millions of active users? Will there be hundreds of clients used by niche groups or 3-4 dominant ones used by the majority?
I couldn't help but notice the parallels between Farcaster clients and Internet browsers.
Today, most of the world uses a few browsers that dominate the market and define the majority of people's experience with the internet (i.e. Chrome, Firefox, IE). But has it always been like this from the start? How come we don't all use different kinds of browsers today?
I went through a quick rabbithole of how the browser market share has changed over time. It's quite fascinating to see how dominant the few players that existed were.
The Browser Wars
This history of browsers can be broken down into 3 simple eras.
First Era: Netscape vs Microsoft
As I mentioned last week in my Netscape post, it was Marc Andreessen, Eric Bina, and Jim Clark who really got the race started to give people an interface to use the web simply.
Mosaic was the very first browser that was coded up in the basement computer science lab of UIUC by Andreessen and his friends. Then that core team eventually partnered with Clark in Silicon Valley and turned their fun side project into one of the most dominant tech businesses of the early 90s.
Less than two years after Netscape launched, they commanded over 80% of the browser market share.
But then, Bill Gates caught on to the internet wave, and things started going downhill from there. Gates was cunning and decided to make Microsoft's Internet Explorer completely free and bundled it with Windows 95.
Very quickly, more and more people started using the terms "internet" and "IE" interchangeably. Microsoft, through their unbreakable PC moat, was quickly eating Netscape's market share by simply adding the web browser as a cherry on top. Gates didn't even need to profit off of the IE, Windows was his main bread and butter.
Second Era: Microsoft vs ???
By the end of the 90s and the early 2000s, Microsoft dominated the browser wars. It's not even funny as to how much command Gates had over the world's experience with the internet. By 2003, almost 95% of people using browsers clicked on the iconic blue "e" to browse the web.
The only honorable mention here is Mozilla around 2009 - they were able to get 30% of the space but it was a short lived victory.
Third Era: Microsoft vs Chrome
Then finally, at the turn of the decade, Google hopped into the ring and Chrome starting going to the moon 🚀 Chrome was a fresh take on the browser after many years of the boring, old Microsoft UI. Primarily, many people switched over because it was considerably faster with the "sandboxing" technique that ran each tab as its own process.
Furthermore, Google used the "App Store" model that Apple introduced with the iPhone a few years prior, and decided to launch the Chrome store. Many developers immediately got excited with this new canvas they could experiment with.
And most importantly, Google made Chrome's underlying code, Chromium, open-source which created a unique brand and build trust with the new ethos of privacy the mainstream media was catching onto.
Simply put:
1990s: Netscape
2000s: Internet Explorer
2010s: Google Chrome
So with this historical context of browsers, how can we think about the future of Farcaster clients? Or really any clients being built on top of open source protocols here on out?
ChatGPT, debug this error!
Based on the browser market share timeline above, the obvious answer would be that there will probably be 1-2 main clients that everyone uses to interact with Farcaster. Maybe the initial one built by the core team - Warpcast - and one other one that catches steam.
But what are the arguments against this answer?
Well I think it's worth mentioning two counter-points:
The speed of software development
It's no surprise that in the last 20 years, the number of resources and people shipping software has dramatically risen. But the key unlock here is pretty recent: using AI copilot. It's a lot easier for an individual to ship software with the help of new AI tooling.
To be clear - I'm not saying everyone and their moms will be making a new Farcaster client. Writing the code is one part of the equation; don't forget that you still have to build a product people want to use.
However, in the next decade, as the AI co-pilot tooling becomes more robust, it's not crazy to think that we might be able to spin up a new client with a simple command in English. As Chris Carella puts it, "in a long enough time horizon we will each have our own bespoke 1/1 client".
This framing makes an interesting case for having the ability to gives all users a completely optimized social media experience at any given time.
The Cost to build a browser is a lot more
It might just be the fact that browsers are a much bigger and bolder project than building a Farcaster client so this comparison only is valid to a certain extent. I didn't realize it, but Dan Romero mentioned a good point - the cost of building a browser costs in the ballpark of hundreds of millions to billions of dollars.
Of course, it's worth noting that the price to produce software has also become significantly cheaper since the 90s. I'm planning to listen to this episode of the Acquired podcast with Brandon Eich, the chief architect of Netscape and Mozilla to better understand why it was so difficult for people to build browsers.
I'm not sure what's going to end up becoming the norm for Farcaster and clients for other crypto products over the next few years, but it'll definitely be interesting to see the way people end up congregating.
My guess is that there will be one or two general use case clients that dominate the web3 social space and then a long list of niche clients that people use for oddly specific situations.
One main point to note regardless of what you think the answer is - just the fact that this is a question at all is incredible! How many different ways can you interact with Twitter? That's right...just one. We're all forced to interact with the so called "public town square" in one standard format that a billionaire change change up whenever.
The last thing I would mention is that though I used Farcaster clients as an example for this post, but the same framework can be applied across any open source protocol being built. For example, how many Zora clients will be there? What about Nouns clients?
There's going to be this sweet spot we hit with the protocol-client relationship and it's going to be interesting to see how AI and cheaper software development change the game.
I also think the answer itself might be changing even in the long run. Heck, even now, almost 30 years since the launch of Netscape, there is still an endless stream of new browsers being worked on. Have you checked out Arc or Brave? It's that wild to think these new browsers might start breaking Chrome's dominant share today.
Comment or reply with what you think is the most probable outcome for the future of the "Farcaster client wars"!
By learning tech history, it's easier to identify tricks & techniques that can help founders approach the idea maze even today.
That's all for today's post - if you enjoyed, I'd love for you to share with your friends in crypto :)
Remember, all posts are collectible as well! Just connect your wallet and mint away.
Also, if you haven't already, please join The Bigger Picture community by hitting the subscribe button below. You can connect your wallet or add your email!
- Loading comments...