In the 1970s, Xerox PARC (Palo Alto Research Center) was a budding hub for innovation, kind of like a miniature version of Bell Labs during the first half of the 20th century. Bell Labs was an innovation hub, and it led to some of the most amazing innovations of the 20th century, not to mention an incredibly robust network of telephones that connected the entire world together in real time.
Robert Metcalfe was working on a new kind of network at Xerox PARC. This wasn’t the vast telephone network Bell Labs had steadily built over half a century, but it was a close cousin.
The existing ARPANET (the early proto-Internet) was great for a small handful of far-away users; it famously connected two cities in California during the so-called Mother Of All Demos, but Metcalfe was interested in solving a different problem altogether.
On May 22nd, 1973, Metcalfe penned a memo he titled “Alto Ethernet.” This addressed the technical challenges so that lots of individuals could hook their computers up locally. Early Alto computers (Xerox’s brainchild) would share files and even printers.
It’s probably fair to think of the Alto Ethernet as the great-grandparent of modern Ethernet, and the Ethernet is like the backbone to the internet. The actual cables and things that connect us together rely on the descendants of Metcalfe’s protocols and solutions.
Users began to jump onto the Ethernet! As the numbers built up, Metcalfe noticed something interesting: the more people who were connected, the faster additional users started hopping on, and the more useful the network became.
This was not a linear relationship, where if you had ten users and added one to the network, you’d get a 10% increase in how useful the network was. Instead, Metcalfe observed that the value of a telecommunications network is proportional to the square of the number of connected users of that network.
Put a little too simply, the more people using a network, the more valuable it becomes. Less simply, with a little nuance: the value is determined by the connections, not the number of people. If there are ten people in a network, I can talk to 9 other people, and those other 9 people can also talk to 9 other people. The number of connections each of us can make is 9, so the use of the number of possible connections is 90 (9 users times 10 possible connections each).
If there are 20 people—twice as many—then every individual can talk to 19 others, so 20 x 19 = 380. If you double the number of users here, the network itself becomes more than four times as valuable.
We have some easy real-life examples in today’s social media networks. I remember first getting onto early proto-social networks during the early 2000s, but sites like Friendster and Orkut just weren’t quite ready for prime time. However, Myspace changed all that for me around 2004, and Facebook followed closely on its heels.
As more users hopped on, the network became more and more valuable. This value increased vastly with time, as advertisers hopped on to cash in. The more users who joined, the more incentives others had to join.
“Are you on Facebook? Let’s connect” came to replace exchanging numbers at a bar. Parents and grandparents gradually made their way to the platform, creating tens of billions of dollars in economic value (and plenty of chaos in the world).
Cryptocurrencies are another area where network effects could play an enormous role. We’ve already seen a huge rise in the value of Bitcoin and ETH over the last decade, as more and more users speculate over how widespread the use of each will be.
Virtual reality is probably a great use case for Metcalfe’s Law, but it’s still waiting on enough connections to be perceived as all that valuable just yet.
My martial arts gyms rely on Metcalfe’s Law, at least to a degree. The number of potential training partners you can have is one enormously important feature of a bigger gym, especially if you might typically have a tougher time finding someone your size at a typical martial arts gym.
What are some other areas (besides social media or a small business) where Metcalfe’s Law and these types of network effects really seem to matter? Can you come up with an example for me today?
Okay, here's the counter position. I don't think, on balance, it's healthy to be connected to that many people. I think this ability for all of us to connect to ever-increasingly numbers of people outside of our communities is actually a huge problem that's making the world worse. (says the person who is at this very moment doing exactly that. Yep, I know.)
Many of us now interact with more people in a single day than past generations would have in a whole lifetime. I think the damage this is doing is incalculable -- to our nervous systems, our ability to focus (I'm writing this comment instead of working on my Beatles podcast...), and perhaps most notably, our ability to connect with our geographical communities, and focus on making those communities better, rather than halfway around the world in a place we'll probably never visit and where we know not a single person in real life (and sometimes not even online.)
How many people donate to causes in faraway places (or political campaigns in states they've never even been to?) and do absolutely nothing to help their local communities?
Example: Fifteen minutes outside of Santa Fe, one of the most "progressive" cities in the US, there are communities suffering from exactly the same problems as many communities in distant third world countries -- lack of access to water (at all, not just clean water) and electricity and healthcare and education. I tried for years, when I lived there, to get the "progressive" citizens of Santa Fe to pay attention to this problem, but they're all are too busy worrying about the problems of people half a globe away to pay any attention at all to the suffering of their neighbors. It's not sexy, you see, compared to saving Tibet or sending donations to Ukraine.
I think all of this connectivity is also wrecking our nervous systems and stressing all of us out. We're now exposed to -- and somehow expected to care about -- every single bad thing happening to anyone anywhere in the world. That's not doable or reasonable, to care about that many things in any useful way, but it's now the definintion of a "good person," to spend lots of energy worrying about more things than we can possibly worry about. It's overstressing our compassion, our nervous systems, our ability to truly do anything meaningful about any of it.
Just because something can be done (in terms of connection) doesn't mean it's a good thing to do it. And yet here we are, on substack. I wonder why it is, that we continue to think that being connected to more people is inherrently better, even when (at least for me) I see that it's not actually working out all that well.
I still miss Friendster, or at least my perception of what it was. In reality, it was probably pretty lacking. But at the time, I didn't get why people ditched it for myspace.