Open Source is bigger than you can imagine
Josh Bressers
Vice President, Security
Anchore
$ whoami
The data
This is all of the data from ecosystems
Versions per month
How big is big?
8.1 million
How big is big?
93 million
Number of versions
35,000
What has 25,000 versions?
Number of versions per package
How do all the ecosystems compare?
What do the various ecosystems look like
How big is this really?
4 million total packages
What if we audited 1000 a day?
That’s 11 years of audits
How big is this really?
4 million total packages
What if we audited 1000 a day?
That’s 11 years of audits
And we’re publishing about 8000 new releases per day
How many people are maintaining these things?
0-50 just npm
But the popular projects aren’t one person!
1.4 Billion
But the popular projects aren’t one person!
Percentiles
95th percentile is 2000 downloads
So let’s look at downloads > 100,000
But the popular projects aren’t one person!
But the popular projects aren’t one person!
OK, but everything over a million downloads is different!
But the popular projects aren’t one person!
But the popular projects aren’t one person!
ONE BEEELLLION!!!
But the popular projects aren’t one person!
Who is doing this?
Credit: Tidelift 2023 Open Source Maintainer Survey
Wait, what about this downloads graph
Zoom and enhance (10,000)
Zoom and enhance (100)
How many packages are more than a year old?
Also CVE growth
NVD is broken
A consortium can fix it!!!
CVE by year
The Kernel
The Kernel
Everyone is being crushed
Are we doomed?
Probably not
What’s the point of it all?
“If I had an hour to solve a problem I'd spend 55 minutes thinking about the problem and 5 minutes thinking about solutions.”
– Albert Einstein
It’s an infinite game
Finite Games: Known players, fixed rules, an agreed-upon objective (e.g. baseball).
Infinite Games: Known and unknown players, changeable rules and the objective to keep the game in play (e.g. the cold war).
Open source is different
There’s nothing wrong with open source, this is how it works
There’s something wrong with what we expect from open source
Questions?