3 ms·

Facebook had what turned out to be an accidentally brilliant way of bootstrapping networks and trust. Fundamental problem: social networks with no-one on them

Facebook had what turned out to be an accidentally brilliant way of bootstrapping networks and trust. Fundamental problem: social networks with no-one on them are not fun and not sticky. Fundamental problem 2: if anyone can read what you're writing, it's not a social network, it's a blog. So you can't just let anyone see what people are writing. The way FB handled that in the early days was that by default, people in the same university network could see your status etc. I don't remember exactly what they could and couldn't do but the point is that they created a semi-privileged circle of people with some access based on a pre-made group that people would be familiar with - fellow students at the same university. That gave people a network from day 1 in what felt like a controlled way. Also, as noted, they started with a desirable closed group that people wanted to belong to.

pdpi 7y ago
Crucially, they had a strategy that worked for that time. Same strategy applied today would simply fail.
sah2ed 7y ago
Facebook started out as a social network where you could get what we today regard as PII (e.g phone numbers of the opposite gender and such) for Harvard students, and its engagement numbers were really good (based on Zuck’s previous experience with the Harvard-only FaceMash). They would later expand to other Ivy league schools — aka networks — one by one, rather than simultaneously, as a way to slow their growth. I imagine an unintended side effect of this growth strategy was the sense of FOMO that it engendered among university students that had heard about but couldn’t join the exclusive/elusive Facebook, unlike the other social networks like Friendster, Hi5 etc. For each network on their expansion list, Zuck & team would furiously scrape the student directory to seed the new network with plausible student info.
erik_seaberg 7y ago
In the early days, you could only friend people at your school because that's who was colocated on your MySQL shard. Fully general addressing of sharded objects was added quite a bit later, about when they opened to non-.edu addresses.

pdpi 7y ago

Crucially, they had a strategy that worked for that time. Same strategy applied today would simply fail.

sah2ed 7y ago

Facebook started out as a social network where you could get what we today regard as PII (e.g phone numbers of the opposite gender and such) for Harvard students, and its engagement numbers were really good (based on Zuck’s previous experience with the Harvard-only FaceMash). They would later expand to other Ivy league schools — aka networks — one by one, rather than simultaneously, as a way to slow their growth. I imagine an unintended side effect of this growth strategy was the sense of FOMO that it engendered among university students that had heard about but couldn’t join the exclusive/elusive Facebook, unlike the other social networks like Friendster, Hi5 etc. For each network on their expansion list, Zuck & team would furiously scrape the student directory to seed the new network with plausible student info.

erik_seaberg 7y ago

In the early days, you could only friend people at your school because that's who was colocated on your MySQL shard. Fully general addressing of sharded objects was added quite a bit later, about when they opened to non-.edu addresses.