World’s Code
Have you ever wondered where all of the code for your favorite websites and apps are stored? Maybe it's on some sort of network of secret databases and servers that are guarded by 7 layers of encryption and firewalls and more security than the Pentagon?
Well, at one point in time, that was probably the case but nowadays, everything is on the cloud as engineers should be able to access the codebase from anywhere in the world at any time. Hearing this, you’re probably inclined to think that all of the world’s code is stored on AWS or maybe Azure given the title of this article. And you wouldn’t be wrong.
Technically, it is stored on Azure but it’s not because of Microsoft, thanks to one of their subsidiaries called GitHub which they paid $7.5 billion for. GitHub is home to code repos from Airbnb, Shopify, Netflix, Robinhood, Reddit, DoorDash, Walmart, and 90% of the Fortune 100.
This translates to control over 100 million developers, 4 million organizations, and 330 million repositories.
Essentially, if you were able to hack into GitHub’s servers, you would get access to the entire world’s code whether it’s financials, social media, games, AI, you name it. As GitHub likes to claim, “This is where the world builds software” and now, it’s all controlled by Microsoft. But how did one company end up getting so much power over it?
Well, it wasn’t always like this. GitHub isn’t some sort of legacy tech company founded in the 80s and 90s. GitHub was only founded 15 years ago in 2008 and it started off as just a regular old bootstrapped startup funded by four ambitious engineers.
Yet somehow, they’ve been able to convince the entire world to jump on board, but is this actually safe and responsible? I mean, we’re essentially risking the entire tech world and really much of the world economy on Microsoft’s security measures.
Well, join me as you take a look back at how Microsoft came to control the entire world’s code and explore whether this is actually in the world’s best interest or just Microsoft’s best interest.
Also Read: How IBM REDHAT Secretly Controls The Worlds Servers
GIT
Speaking of Microsoft’s domination, with over $100 billion in cash and liquid investments, Microsoft wipes the floor with other Fortune 500 companies when it comes to cash reserves. Most of these liquid investments are in corporate and treasury bonds. As rates rose over the past year, Microsoft leaned into bonds and locked in high yields from the government and other top corporations.
So while Microsoft profits from higher rates, you can too by investing in bonds through Silo. To understand how Microsoft came to dominate this world, we first have to understand how code used to be stored and distributed before the 2000s because it was really bad.
I mean, just think about it. When you have thousands of coders who are working on the same codebase and making millions of edits to add new features and fix bugs, you’re naturally going to run into a lot of issues. Joe’s bug revisions are going to end up breaking Bob’s new feature.
John’s new efficient algorithm might’ve forgotten a critical edge case for Mary’s legacy functionality and so on and so forth. The bottom line is that getting code from thousands of engineers to work together in unison is a very difficult task in itself, and for the longest time, the solution for this issue was simply brute force.
Each tech company would have its own database, servers, and revision control systems, and dozens of code librarians and senior engineers would be hired to simply maintain a healthy codebase. These guys wouldn’t write their own code or algorithms or any of that. Instead, their full-time job was to review each and every change to ensure that none of them would break any existing feature or conflict with parallel changes.
Also Read: Why AI Will Never Replace Our Jobs: The Human Edge in the Age of Automation
Honestly, this seems like a pretty tedious and annoying job but it was also very much critical and necessary until a young man named Linus Torvalds decided to step in. Just kidding, Linus didn’t care about this internal software engineering issue. He was looking to fry much bigger fish and compete directly with Microsoft.
If you haven’t figured it out yet, Linus was the creator of Linux and that’s where he spent the first 15 years of his professional life. But eventually after Linux grew to be much bigger than just himself, he would turn his attention to one of his biggest frustrations of software engineering: version control.
Linus was more aware of this issue than anyone else as he was the head of the biggest open source project of all time. At least with companies, you can be sure that every engineer is skilled and has the companies best interest in mind atleast to some degree because well they’re getting paid.
With open source software, on the other hand, literally anyone can contribute, so it was up to people like Linus to ensure that only the positive changes made their way onto the repo. This was always a pain in the butt for Linus but he put up with it as he was more focused on the bigger picture.
But on one fateful night in 2005 when working on the Linux Kernel, he would reach his wits end with crappy version control. He vowed right then and there that he was gonna solve this issue once and for all. And with that, he would go offline for 2 weeks, create a new revolutionary version control system from scratch, and call it Git.
Honestly, it’s pretty funny to think that one of the biggest shortfalls within the software engineering industry could’ve been solved or at least largely improved with just 2 weeks of focused work. But for some reason, no one had decided to take on this project for decades.
Talk about procrastination. Linus would of course turn around and implement Git into Linux’s repo which immediately gave Git a strong reputation within the open source community. But there was still a long way to go in terms of making Git mainstream.
Also Read: Why People Are Switching From Android to iOS :The Future of Smartphones
GITHUB
You’re probably wondering, what in the world did Linus do within just 2 weeks that was so revolutionary? Well, he largely reduced the work of code librarians aka the gatekeepers. Before Git, engineers were directly working on the central codebase and this is why having highly qualified gatekeepers was so important.
These gatekeepers were in charge of making sure that only high quality code made its way onto the repo but this also meant that engineers were spending a lot of their time just waiting on gatekeepers instead of writing code.
Ironically, this is still the case at most big tech companies but for different reasons like bureaucracy and varying priorities. Back in the day, the issue was mainly due to a lack of good coding infrastructure. Git, however, largely solved this.
Instead of having coders work directly on the central repository, Git allowed coders to work on a local version of the repo. This meant that coders could test any changes they wanted locally without the oversight of gatekeepers or jeopardizing the central repo. As such, coders were able to better troubleshoot and validate their code before they ever sent it to the gatekeepers to verify and merge with the central repo.
This was a revolutionary development but Git was by no means perfect. For starters, there was no frontend to Git. Everything was handled through the terminal making it difficult to visualize the history of changes and modifications to a given repo.
Moreover, sharing code between engineers still wasn’t all that streamlined or polished, and this is where 4 engineers named Tom Preston Werner, Chris Wanstrath, PJ Hyett, and Scott Chacon would step in. In 2008, these 4 would get to work on putting a face to Git making it easier than ever to share, fork, modify, push, and merge code.
For the first time, every developer on a project could easily view the entire code history, identify who made what changes, figure out where things went wrong, and fix conflicts and production issues with ease. If a lot of those terms went over your head, they basically created an insane code management platform based on Git and they called it Logical Awesome LLC.
Unfortunately, though, that name didn’t stick and they would end up calling it GitHub. Almost immediately, GitHub was a success on its own. It didn’t need any venture capital or funding or loans or any of that. In fact, the founders were able to manage all of the expenses of Git by themselves for the first 4 years. After all, it’s not that expensive to store and distribute a bunch of text.
One of the first companies to embrace Git was none other than Google in 2008. This was during the peak of their obsession with opensource software like Android and Chromium. They knew that if they really wanted to appeal to the opensource community, they needed to use the same tools which in this case was Git.
Google would eventually move away from Git to its own version of a controlling tool called Piper, but that doesn't really matter because Google had already put Git and as a result Github on the mainstream radar. By 2010, Github was hosting 1 million repos, and by 2013, they were hosting 10 million repos.
Most tech companies were just happy to finally have a modern code management platform, but some wanted a piece of the pie. This, of course, included VCs who poured hundreds of millions into GitHub but it also included Microsoft.
Also Read: Why Amazon Is Happy They Are Getting Sued :How It Could Benefit Them
MICROSOFT TAKES CONTROL
Microsoft just didn’t go straight for the acquisition. Instead, they started off as just regular old GitHub users and supporters. Starting in 2012, they would move all of their open source projects, dev tools, and most of their documentation onto GitHub. This did wonders for GitHub. It’s one thing to support a platform.
It’s another to put much of your own code on that platform, and at this point, both Google and Microsoft had taken the plunge. Let’s just say, it was just a matter of time until everyone else followed suit. Microsoft watched all of this playout from the sidelines for several years before finally making their move in 2018.
The offer was $7.5 billion for the entire company and frankly this was a deal that the founders couldn’t turn down as it would make them billionaires. But with that being said, the deal wasn’t really hostile. You see, Microsoft wasn’t trying to takeover and impose their control over the company. All they really wanted was a star investment and cloud partner and that’s exactly how they’ve treated it.
If you don’t believe me, just take it from GitHub’s CEO himself. “Microsoft has not forgotten why we did the deal in the first place and what the important pillars of the deal are. The first and foremost principle is to put developers first. And that's what we do every day.” That’s a far cry from how the founders of Instagram and WhatsApp feel about Zuckerberg but that’s a whole other story.
Anyway, as you can see, things have worked out great for both GitHub and Microsoft, but what about everyone else? Is it healthy and safe for one company to have so much power? Well, if the question is healthy, the answer is absolutely not. It is not healthy for one company to control the codebase of 90% of the fortune 100.
They absolutely need competition and maybe even regulators but that’ll likely never happen because GitHub is a background company that most people don’t even know or care about. So, as long as they continue lurking in the background, they shouldn’t have many issues with regulators. But while this system is definitely not healthy, it is safe.
Given GitHub’s business, they’re naturally a target for hackers. A few years ago, for example, they were subject to the largest DDoS attack ever recorded. But we don’t really have to worry about this because GitHub’s customers know what they’re doing.
Here’s the thing, 99% of the code in the world doesn’t matter, not in the sense that it’s not important but in the sense that it’s not proprietary. Many of the largest projects in the world, like Chrome and Android, are already opensource. And even for projects that aren’t opensource, the code base doesn’t really matter.
Just because you have Netflix, YouTube or Reddit’s codebase doesn't mean you’ll even have a chance at competing against them. For most tech companies, their secret source is their user base, their marketing, their partnerships, their brand presence, and so on.
There’s really nothing proprietary about making API calls or storing data or handling errors, which is what the vast majority of the world’s code does. So companies aren’t exactly worried about their codebases getting leaked, though they would of course prefer not to go down that road.
There are very few projects and companies that actually deal with proprietary code where the code truly is the hero. You can bet that these projects are not stored on GitHub. They’re stored in secret databases with insane levels of security like in the good old days. But aside from these few exceptions, virtually everything is stored on GitHub and that’s how Microsoft secretly controls every app in the entire world.
Also Read: How Microsoft Was Blackmailed into Acquiring Nokia: Unveiling the Inside Story
